Other Matters Then Pend- Ing; and Such Emergency Matters As May Be Submitted by the Governor

Home , Abel Herrero, Blake Farenthold, Burt Solomons, Carlos Uresti, Charlie Gonzalez, Ciro Rodriguez

No. 17-586

IN THE Supreme Court of the United States ———— GREG ABBOTT, in his official capacity as Governor of Texas, et al., Appellants, v. SHANNON PEREZ, et al., Appellees. ———— On Appeal from the United States District Court for the Western District of Texas ———— MOTION TO DISMISS OR AFFIRM SUPPLEMENTAL APPENDIX ————

JOSÉ GARZA RENEA HICKS MARTIN GOLANDO Counsel of Record GARZA GOLANDO LAW OFFICE OF MORAN, PLLC MAX RENEA HICKS 115 E Travis St., Ste. 1235 P.O. Box 303187 San Antonio, TX 78205 Austin, TX 78703 (512) 480-8231 Counsel for Appellee [email protected] Mexican American Legislative Caucus, Texas MARC E. ELIAS House of Representatives BRUCE V. SPIVA ARIA C. BRANCH DAVID RICHARDS PERKINS COIE LLP RICHARDS, RODRIGUEZ & 700 13th St. NW, Ste. 600 SKEITH, LLP Washington, DC 20005 816 Congress Ave., Ste. 1200 Austin, TX 78701 ABHA KHANNA PERKINS COIE LLP Counsel for Appellees 1201 Third Ave., Ste. 4800 Shannon Perez, et al. Seattle, WA 98101 Counsel for Appellees Eddie Rodriguez, et al. Additional Counsel Listed on Inside Cover November 20, 2017

WILSON-EPES PRINTING CO., INC. – (202) 789-0096 – WASHINGTON, D. C. 20002 ANITA S. EARLS J. GERALD HEBERT ALLISON J. RIGGS J. GERALD HEBERT, P.C. SOUTHERN COALITION FOR 191 Somervelle St., No 405 SOCIAL JUSTICE Alexandria, VA 22304 1415 West Highway 54 MARK P. GABER Ste. 101 439 Elm St. NW Durham, NC 27707 Washington, DC 20001 ROBERT NOTZON JESSICA RING AMUNSON LAW OFFICE OF JENNER & BLOCK LLP ROBERT S. NOTZON 1099 New York Ave. NW 1507 Nueces St. Ste. 900 Austin, TX 78701 Washington, DC 20001 VICTOR L. GOODE GERALD H. GOLDSTEIN NAACP GOLDSTEIN, GOLDSTEIN, 4805 Mt. Hope Dr. & HILLEY Baltimore, MD 21215 310 S. St. Mary’s St. Counsel for Appellees the San Antonio, TX 78205 Texas State Conference of DONALD H. FLANARY, III NAACP Branches, et al. FLANARY LAW FIRM LUIS R. VERA, JR. 1005 South Alamo LAW OFFICES OF LUIS San Antonio, TX 78210 ROBERTO VERA, JR. JESSE GAINES 1325 Riverview Towers P.O. Box 50093 111 Soledad Fort Worth, TX 76105 San Antonio, TX 78205 Counsel for Appellees Counsel for Appellees Margarita Quesada, LULAC, et al. et al. GARY L. BLEDSOE ROLANDO L. RIOS POTTER BLEDSOE, LLP LAW OFFICES OF 316 West 12th St. ROLANDO L. RIOS Ste. 307 115 E. Travis St., Ste. 1645 Austin, TX 78701 San Antonio, TX 78205 Counsel for Appellees Counsel for Appellee Congressmembers Eddie Congressman Henry Bernice Johnson, Sheila Cuellar Jackson Lee and Al Green and NAACP Intervenor JOAQUIN G. AVILA Howard Jefferson P.O. Box 33687 Seattle, WA 98133 Counsel for Appellee Mexican American Legislative Caucus

1a SUPPLEMENTAL APPENDIX IN THE UNITED STATES DISTRICT COURT FOR THE WESTERN DISTRICT OF TEXAS [Filed 03/10/17] ———— SA-11-CV-360 ———— SHANNON PEREZ, et al. v. RICK PERRY, et al. ———— FACT FINDINGS - GENERAL AND PLAN C185 ———— Before Circuit Judge SMITH, Chief District Judge GARCIA, and District Judge RODRIGUEZ Circuit Judge Smith dissenting XAVIER RODRIGUEZ, District Judge and ORLANDO L. GARCIA, District Judge: BACKGROUND and TIMELINE 1. These fact findings are based on evidence presented at all phases of this litigation. Citations to the record from the September 2011 trial are shown as “Tr”; citations to the record from the July 2014 trial are shown as “TrJ”; citations to the record from the August 2014 trial are shown as “TrA.” 2. Any finding of fact herein that also constitutes a conclusion of law is adopted as a conclusion of law. 2a 3. The Court has considered the stipulated facts submitted in docket number 277 at 15-25 (joint pre- trial order), docket no. 302 (stipulation of facts regarding plaintiffs between Quesada Plaintiffs and Defend- ants), and docket no. 304 (stipulation between Texas Latino Redistricting Task Force Plaintiffs and Defend- ants regarding the King Ranch). 4. The Court has determined the race and party of members of the Texas House of Representatives in part from a chart prepared by Defendants, exhibit D-157. 5. The Court has taken judicial notice of election returns available on the Texas Secretary of State’s website. 6. The Court has taken judicial notice of certain census and American Community Survey (“ACS”) data set forth in docket no. 1085. See 6/20/14 text order granting motion as to items 1-8. 7. The Court has taken judicial notice of election returns in Nueces County and Kleberg County as set forth in docket no. 1169. TrJ2160-61. 8. Every ten years, under 2 U.S.C. § 2a, the President of the United States must transmit to Con- gress a statement showing the number of persons in each state and the number of representatives in the United States House of Representatives to which the state is entitled. These figures are tabulated according to the federal decennial census. Historically, there has never been a completely accurate census in the United States. (Stipulated.) 9. The Texas Legislature meets in regular session on the second Tuesday in January of each odd-num- bered year. TEX. CONST. art. III, § 5; TEX. GOV’T CODE 3a § 301.001; TrJ1650 (Hochberg). The Texas Constitu- tion limits the regular session to 140 calendar days. TEX. CONST. art. III, § 24(b). It further provides, “When convened in regular Session, the first thirty days thereof shall be devoted to the introduction of bills and resolutions, acting upon emergency appropriations, passing upon the confirmation of the recess appointees of the Governor and such emergency matters as may be submitted by the Governor in special messages to the Legislature. During the succeeding thirty days of the regular session of the Legislature the various committees of each House shall hold hearings to consider all bills and resolutions and other matters then pending; and such emergency matters as may be submitted by the Governor. During the remainder of the session the Legislature shall act upon such bills and resolutions as may be then pending and upon such emergency matters as may be submitted by the Governor in special messages to the Legislature.” TEX. CONST. art. III, § 5(b). The Texas Constitution mandates that the Legislature, “at its first regular session after the publication of each United States decennial census, apportion the state into senatorial and representative districts.” TEX. CONST. art. III, § 28. The Texas House rules require five days notice for a public hearing on legislation in a regular session. TrJ1650 (Hochberg); D-669. 10. If a Texas House or Senate map is not passed in the first regular session following the census, the Legislative Redistricting Board (“LRB”), which is “composed of five (5) members, as follows: The Lieutenant Governor, the Speaker of the House of Representa- tives, the Attorney General, the Comptroller of Public Accounts and the Commissioner of the General Land Office,” will draw the map. TEX. CONST. art. III, § 28. In 2011, the LRB was composed of five Republicans. 4a TrJ39 (Veasey). Although it may, the LRB does not have to hold public hearings, review the legislative record, or solicit input from members. TrJ39-40 (Veasey). 11. The Legislature may draw a congressional map during a regular or special session. Special sessions last 30 days. TrA1087 (Hunter); TrA1557 (Hanna). Multiple special sessions may be called. 12. Since 1876, Article III, Section 26 of the Texas Constitution, which includes the County Line Rule, has provided as follows: The members of the House of Representatives shall be apportioned among the several counties, according to the number of population in each, as nearly as may be, on a ratio obtained by dividing the population of the State, as ascertained by the most recent United States census, by the number of members of which the House is composed; provided, that whenever a single county has sufficient population to be entitled to a Representative, such county shall be formed into a separate Representa- tive District, and when two or more counties are required to make up the ratio of representation, such counties shall be contiguous to each other; and when any one county has more than sufficient population to be entitled to one or more Representatives, such Repre- sentative or Representatives shall be apportioned to such county, and for any surplus of population it may be joined in a Representa- tive District with any other contiguous county or counties.

5a 1990s 13. According to the 1990 Census, the State of Texas had a total population of 16,986,510 persons, consisting of 60.6% non-Hispanic white persons, 25.5% Hispanic persons, 11.6% non-Hispanic black persons, and 1.8% non-Hispanic Asian persons. Docket no. 1085, Ex. 2 at 2 (judicially noticed per text order 6/20/14). Texas had a voting-age population of 12,150,671 persons, including 64.4% non-Hispanic white persons, 22.4% Hispanic persons, 11% non-Hispanic black persons, and 1.8% non-Hispanic Asian persons. Docket no. 1085, Ex. 2 at 4 (judicially noticed per text order 6/20/14). 14. In 1991, the Texas House Redistricting Com- mittee (“HRC”) held two public hearings, and the Senate State Affairs Committee held one public hearing on the Texas House redistricting bill, H.B. 150. D-285 (no obj). The bill was read a first time (referred to House committee) on January 24, 1991, and the HRC held public hearings on April 15 and May 18 (the posting rule was suspended on May 17). The bill was read a second time on May 21 and amendments were considered. It was read the third time on May 22, and was referred to the Senate State Affairs Committee on May 22. The Senate State Affairs Committee held a public hearing on May 23, and the bill was read a second and third time in the Senate on May 24, and passed the Senate the same day. The bill was signed in the Senate and the House on May 26. D-285. Therefore, the 72nd Legislature passed a House map in regular session. The plan was challenged in court, and a new House plan was adopted in a January 1992 special session. Litigation over the House plan continued through 1997. 6a 15. Texas’s population increase, largely in urban minority populations, entitled it to three additional congressional seats. Bush v. Vera, 517 U.S. 952, 956-57 (1996). The Texas Legislature promulgated a redistricting plan that created CD30, a new majority- African-American district in Dallas; created CD29, a new majority-Hispanic district in and around Houston; and reconfigured CD18 in Houston to make it a majority-African-American district. Id. at 957. The DOJ precleared the plan, and it was used in the 1992 and 1994 elections. Six Texas voters challenged the plan as including racially gerrymandered districts in violation of the Fourteenth Amendment, and the three-judge court held CD18, CD29, and CD30 unconstitutional. Vera v. Richards, 861 F. Supp. 1304 (1994). The Supreme Court affirmed in Bush v. Vera, 517 U.S. 952 (1996), finding that traditional redistricting principles were subordinated to race and that the three new minority-majority districts did not survive strict scrutiny because they were not compact and therefore not required by § 2 of the VRA. The three- judge court then imposed an interim plan in Vera v. Bush, 933 F. Supp. 1341 (1996), redrawing the boundaries for CD18, CD29, and CD30 and portions of other nearby districts. 2000-2001 16. Between 1990 and 2000, the state grew by about 3.9 million, and three million of that growth was minority. Tr862 (Murray). Hispanics made up about 60% of the growth and African-Americans almost 12%. Joint Expert Ex. E-4 (Murray report) at 8. The State began to see a pattern of slowing Anglo growth and faster minority growth. Tr863 (Murray). 17. According to the 2000 Census, the total population of 20,851,820 included 10,933,313 non-Hispanic 7a white persons (52.4%), 6,669,666 Hispanic persons (32.0%), 2,399,083 non-Hispanic black persons (11.5%), and 594,932 non-Hispanic Asian persons (2.8%). Docket no. 1085, Ex. 2 at 1-3 (judicially noticed per text order 6/20/14); US-638.1 Texas had a voting-age population of 14,965,061 persons, including 8,426,166 non-Hispanic white persons (56.3%), 4,282,901 Hispanic persons (28.6%), 1,639,173 non-Hispanic black persons (11.0%), and 437,215 non-Hispanic Asian persons (2.9%). Docket no. 1085, Ex. 2 at 4-6 (judicially noticed per text order 6/20/14); US-638. Texas had a citizen voting-age population (“CVAP”) of 13,299,845 persons, including 8,305,993 non-Hispanic white persons (62.5%), 2,972,988 Hispanic persons (22.4%), 1,590,832 non- Hispanic black persons (12.0%), and 225,374 non- Hispanic Asian persons (1.7%). Docket no. 1085, Ex. 3 at 7-11 (judicially noticed per text order 6/20/14). 18. In 2000 and before, the Census Bureau collected citizenship information on the decennial census “long form,” so there was good citizenship data that included smaller geographic areas available for redistricting. Tr1031 (Murray); Tr1099 (Ansolabahere). After 2000, the Census Bureau no longer collected citizenship data as part of the decennial census, instead switching to an ongoing, smaller, annual survey called the Amer- ican Community Survey (“ACS”). The ACS began collecting long-form-type information throughout the decade rather than only once every ten years. D-323. Although under the umbrella of the Census Bureau, the ACS is not part of the decennial census. Tr1033 (Murray); Tr1673 (Rives). The ACS is collected by full-

1 There are some discrepancies in the population numbers among the various exhibits in this case (see, e.g., Quesada-409A, Joint Expert Ex. E-5 (Martin report) at 3). The discrepancies are not material. 8a time professional staff using multimodal data collec- tion methods. Tr181 (Chapa). They survey 250,000 households a month, and the sample is designed to be accumulated over time, so it provides more current data during the decade after a census. Tr181 (Chapa). 19. While the census long form sampled about 18 million housing units, the annual ACS samples only about 3 million addresses. D-323; Tr1673 (Rives); Tr181 (Chapa). Further, the ACS is not a tabulation, but provides estimates about population characteristics. Tr1672-73 (Rives). The ACS traded precision for more frequent updates. Tr1675 (Rives). The 1-year estimates are statistically reliable for areas of 65,000 or larger. Tr1675 (Rives); D-323. For areas between 20,000 and 65,000, there are 3-year ACS estimates. Id. “[T]he ACS needs to combine population or housing data from multiple years to produce reliable numbers for small counties, neighborhoods, and other local areas.” D-323. The ACS therefore provides 5-year estimates for smaller areas such as census tracts and block groups. D-323. Even with aggregated data, block group estimates may contain large margins of error. Joint Expert Ex. E-1 (Chapa report) at 10. ACS data were first available in 2006 in the form of 1-year estimates for 2005, and 1-year estimates have been available every year since then. 20. In the 2001 redistricting cycle for the Texas House of Representatives, the 77th Texas Legislature failed to pass a map, and so the LRB drew a plan. TrJ1927 (Bruce); PL-229. The LRB’s plan (01289H) was submitted to the DOJ for preclearance pursuant to the VRA in August 2001. PL-229. The DOJ objected under § 5 of the VRA, specifically finding retrogression in Bexar County, South Texas (HD35 and HD38), and West Texas (HD74). PL-225; D-326. With regard to the 9a existing benchmark plan, the DOJ’s objection letter listed the number of combined majority-minority districts based on total population and based on voting age population, and, with regard to majority-minority voting age population (“VAP”), listed the number of districts with a majority Hispanic VAP, majority Black VAP, combined minority-majority VAP, and Spanish- Surname Voter Registration (“SSVR”) majority. 2 Id. The letter continued, “An initial issue arises as to the appropriate standard for determining whether a district is one in which Hispanic voters can elect a candidate of choice. The State of Texas has provided, and accepted as a relevant consideration, Spanish- surnamed registered voter data as well as election return information and voting age population data from the census. We agree with the State’s assessment, although we also consider comments from local individuals familiar with the area, historical election analysis, analysis of local housing trends, and other information intended to create an accurate picture of citizenship concerns. Campos v. Houston, 113 F.3d 554, 548 (5th Cir. 1997).” PL-225; D-326. The letter concluded that the LRB House plan would lead to retrogression in the position of minorities “with respect to their effective exercise of the electoral franchise by causing a net loss of three districts in which the minority community would have the opportunity to elect its candidate of choice.” Id. It said, “Although there is an increase in the number of districts in which Hispanics are a majority of the voting age population, the number of districts in which the level of Spanish surnamed registration (SSRV) is more than 50 percent

2 This metric is referred to both as SSVR (Spanish Surname Voter Registration) and SSRV (Spanish Surnamed Registered Voters). 10a decreases by two as compared to the benchmark plan. Moreover, we note that in two additional districts SSRV has been reduced to the extent that the minority population in those districts can no longer elect a candidate of choice. In the State’s plan these four reductions are offset by the addition of a single new majority minority district – District 80 – leaving a net loss of three.” Id. DOJ noted that when Bexar County went from 11 to 10 districts, the State unnecessarily eliminated a Latino opportunity district, and DOJ rejected the State’s argument that there were sufficient new Latino opportunity districts to offset the loss. DOJ also objected to retrogression in HD35, HD38 in Cameron County, and HD74 in West Texas. Id. In HD35, the SSVR dropped to 50.2% from 55.6% and the Hispanic incumbent was paired with an Anglo incumbent in a district that was 58% of the Anglo incumbent’s former district (and thus assumed to favor the Anglo incumbent). In HD38, SSVR was reduced from 70.8% to 60.7% and over 40% of its core population was removed, 90% of whom were Hispanic, and replaced with population that was 45% non-minority. Despite remaining a majority-Hispanic-VAP district (69.6%), DOJ concluded that it was doubtful that Hispanics would be able to elect their candidate of choice. DOJ also noted that districts surrounding HD35 and HD38 had unnecessarily high SSVR levels exceeding 80%, meaning the reductions in HD35 and HD38 were avoidable had the State refrained from packing Hispanic voters into adjacent districts. DOJ also noted that member- constituent relations and communities of interest would be disrupted at a disproportionately higher rate than other areas of the State. In HD74, DOJ concluded that the SSVR reduction from 64.5% to 48.7% meant that Hispanic voters would lose the opportunity to elect their candidate of choice, and that changes to this 11a district were unnecessary given that it was close to ideal population. Id. 21. After a trial considering the DOJ’s objections and the parties’ contentions, on November 28, 2001, the district court for the Eastern District of Texas in Balderas v. Texas ordered into effect a new House plan. PL-228 (judgment); PL-229 (opinion). The Balde- ras court’s order found no violation of § 2 of the VRA in the failure to create Latino districts or coalition districts (as to the coalition districts, the court stated that it was not convinced that the different minority groups voted cohesively). The court also found no violation of the Equal Protection Clause based on intentional discrimination. The court’s plan did remedy the DOJ’s retrogression concerns, including the loss of a Latino district in Bexar County, as well as the reduction of Latino voting strength in districts 35, 38, and 74. Balderas v. Texas, No. 6:01-CV-158, 2001 WL 34104833 (E.D. Tex. Nov. 28, 2001), summarily aff’d, 536 U.S. 919 (2002). The court reconstituted a seventh majority-Latino district in Bexar County and equal- ized the SSVR (of at least 55%) across all seven Latino districts. The court modified HD38 and increased its SSVR to 73.4%. It also reconfigured HD74 and raised its SSVR to 54% while preserving county lines as required by the Texas Constitution. In HD35, the court eliminated the pairing and left the district with no incumbent and an SSVR of 51.5%. This plan is the benchmark plan for this litigation. It is referred to as Plan H100. 22. Following the 2000 census, Texas was apportioned 32 seats in the U.S. House of Representatives, an increase of two. The Texas Legislature failed to enact a plan during the 2001 session, and the Eastern District of Texas ordered into effect Plan 1151C on 12a November 14, 2001. See D-654 (map); Balderas v. Texas, No. 6:01-cv-158 (E.D. Tex. Nov. 14, 2001), summ. aff’d 536 U.S. 919 (2002). 2003 - 2006 23. In 2003, after Republicans gained control of both the Texas House and Senate, the 78th Texas Legislature drew a new map for the United States House of Representatives. Plan 01374C was enacted in the third special session. The plan was upheld by a three-judge court in the Eastern District of Texas. On appeal, the Supreme Court found that the State violated § 2 of the VRA in its configuration of CD23 in West Texas. The Court noted, “After the 2002 election, it became apparent that District 23 as then drawn had an increasingly powerful Latino population that threatened to oust the incumbent Republican, Henry Bonilla.” LULAC v. Perry, 548 U.S. 399, 423 (2006). In redrawing CD23, the Legislature dropped the Latino share of the citizen voting age population from 57.5% to 46%. The Supreme Court stated, “it is evident that the second and third Gingles preconditions—cohesion among the minority group and bloc voting among the majority population—are present in District 23.” Id. at 427. It further found that “the Latino majority in old District 23 did possess electoral opportunity protected by § 2.” Id. Although the old CD23 was a Latino opportunity district, the new CD23 unquestionably was not. Id. The Court noted, “[T]he new plan divided Webb County and the city of Laredo, on the Mexican border, that formed the county’s population base. Webb County, which is 94% Latino, had previously rested entirely within District 23; under the new plan, nearly 100,000 people were shifted into neighboring District 28. The rest of the county, approximately 93,000 people, 13a remained in District 23. To replace the numbers Dis- trict 23 lost, the State added voters in counties comprising a largely Anglo, Republican area in central Texas. In the newly drawn district, the Latino share of the citizen voting-age population dropped to 46% [from above 50%].” Texas argued that it offset the loss of CD23 by creating the new CD25, but the Supreme Court concluded that the new CD25 was not a required § 2 district because it was not compact, and thus could not offset the loss of CD23. Id. at 430-31. 24. The Supreme Court remanded to the Eastern District of Texas, which ordered into effect a new congressional plan on August 4, 2006. LULAC v. Perry, 457 F. Supp. 2d 716 (E.D. Tex. 2006); PL-230 (order adopting plan 1438C); PL-231 (opinion); D-648 (map). The court’s plan redrew districts 15, 21, 23, 25 and 28. The court drew CD23 to make it an effective Hispanic opportunity district, based on its Hispanic Citizen Voting Age Population (“HCVAP”) numbers and its election performance, and re-united Webb County and placed it in CD28. The court rejected configurations that would place Webb County in CD23 because of the “clockwise” consequences in the map, and found that placing Webb County in CD28 allowed the South Texas districts to be more compact. This plan is the benchmark congressional plan for this litigation. It is referred to as Plan C100. 25. In the 2006 special November elections, CD23 incumbent Bonilla received 48.6% of the vote and Ciro Rodriguez (Hispanic, Democrat) received 19.86% of the vote (there were six Democrat candidates, and Rodriguez received the highest number of votes). D- 422. In the special runoff election, Rodriguez prevailed with 54.28% of the vote, defeating incumbent Bonilla. D-421. 14a 2008 26. The 2008 general election was considered good for minority turnout and minority voters. It was also considered a good election year for Democrats, who elected President Obama. Joint Expert Ex. E-2 (Kousser report) at 73-76. 27. CD23 incumbent Ciro Rodriguez (Hispanic, Democrat) won re-election with 55.76% of the vote. CD27 incumbent Solomon Ortiz (Hispanic, Democrat) won re-election with 57.95% of the vote. In El Paso’s HD78, Joe Moody (Hispanic, Democrat) defeated Dee Margo (Anglo, Republican) 51.53% to 45.11%. In HD117 in Bexar County, incumbent David McQuade Leibowitz (Democrat) defeated John Garza (Hispanic, Republican) with 57.03% of the vote. In Nueces County, HD33 incumbent Solomon Ortiz, Jr. (Hispanic, Demo- crat) defeated Raul Torres (Hispanic, Republican) 59.04% to 35.53%. HD34 incumbent Abel Herrero (Hispanic, Democrat) defeated Connie Scott (Anglo, Republican) 53.14% to 46.86%. However, in HD32, Todd Hunter (Anglo, Republican) defeated incumbent Juan Garcia (Hispanic, Democrat) 50.13% to 46.80%. 2009 28. On April 16, 2009, the HRC of the 81st Legisla- ture held a public hearing on redistricting. D-116 at 2. The members of the committee at that time were: Chairman Jones, Vice-Chairman Villarreal, Reps. Hilderbran, Merritt, Peña, Pickett, T. Smith, Veasey, Alvarado, Deshotel, Eissler, Harless, Herrero, Keffer, and Pitts. Testimony was taken on H.B. 104 (relating to the reapportionment of state legislative, congressional, and judicial districts and the creation, function, and duties of the Texas Redistricting Commission) and H.J.R. 19 (proposing a constitutional amendment 15a establishing the Texas Redistricting Commission to establish legislative and congressional districts and revising constitutional redistricting procedures). 29. On September 30, 2009, the HRC met to hear testimony on the upcoming 2010 Census, and heard a presentation from the U.S. Census Bureau. D-116 at 6. Witnesses from MALDEF, NAACP, and LULAC testified on the upcoming census. 2010 30. According to the 2010 United States Census, the Texas population was 25,145,561, including 11,397,345 non-Hispanic white persons (45.3%) (a decrease from 52.4% in 2000), 9,460,921 Hispanic persons (37.6%) (an increase from 32% in 2000), 2,975,739 non- Hispanic black persons (11.8%), and 1,027,956 non- Hispanic Asian persons (4.1%). Docket no. 1085, Ex. 4 at 1-3 (judicially noticed per text order 6/20/14).3 No single ethnic or racial group constitutes a majority of the Texas population. (Stipulated.) 31. Between 2000 and 2010, there was a total population increase of 4,293,741 (or 20.6%) from the 2000 Census. Joint Expert Ex. E-1 (Chapa report) Table 1; Joint Expert Ex. E-5 (Martin report) at 3. Anglo growth slowed and there was virtually none by the end of the decade. Tr866 (Murray). The Anglo population growth rate was 4.2% (a total of 464,032 people). Joint Expert Ex. E-5 (Martin report) at 3. African-American growth accelerated and Hispanic growth continued unabated, both exceeding Anglo

3 The parties have stipulated to the percentages. Again, there are some discrepancies among the exhibits concerning the numbers of African-American and Asian persons, but they are not material (see, e.g., Quesada-409B, Joint Expert Ex. E-5 (Martin report) at 3). 16a growth. Tr866-67 (Murray). The Hispanic growth rate was 41.8%; the African-American growth rate was 22.1%; the Asian growth rate was 71.1%. Joint Expert Ex. E-5 (Martin report) at 3. Hispanic total population, HVAP, and HCVAP are growing rapidly. Tr189 (Chapa). 32. According to the 2010 Census, the Texas Latino population increased by 41.8% over the previous ten years and accounted for 65% of the State’s total population growth. (stipulated). The Hispanic population of Texas grew to 9,460,921 from 6,669,666 in the 2000 Census (an increase of 2,791,255 people). (Stipulated.) 33. According to the 2010 Census, the State of Texas has a voting-age population of 18,279,737, including 9,074,684 non-Hispanic white persons (49.6%), 6,143,144 Hispanic persons (33.6%), 2,102,474 non- Hispanic black persons (11.5%), and 758,636 non- Hispanic Asian persons (4.2%). Docket no. 1085, Ex. 4 at 4-6 (judicially noticed per text order 6/20/14). 34. On February 10, the HRC of the 81st Legisla- ture held a hearing to take invited testimony only relating to the upcoming 2010 Census and population estimates. D-116 at 19. Clare Dyer and David Hanna of the Texas Legislative Council (“TLC”) and Gabriel Sanchez from the U.S. Census Bureau testified. Id. at 22. The TLC is a statutorily created, nonpartisan agency that provides legal, administrative, and technical support to the Texas LEGISLATURE. TEX. GOV’T CODE § 323.001 et seq. Its services are available to all legislators. TrJ1145-47 (Hanna); TrJ226 (Dyer). 35. On June 2, 2010, the HRC held a public hearing with the House Committee on Judiciary and Civil Jurisprudence at the State Capitol extension; only invited testimony was permitted. D-116 at 39; US- 17a 272B. The committee heard testimony from Dyer and Hanna of the TLC and Jeffrey Jordan from the Texas State Data Center. D-116 at 42. 36. The 81st Texas Legislature conducted field hearings on the subject of redistricting throughout Texas from June to December 2010. Most of the interim field hearings were joint hearings with the House Judiciary and Civil Jurisprudence Committee. TrA1064 (Hunter). Then-Chair of the HRC, Delwin Jones, asked Hunter, Chair of the Judiciary and Civil Jurisprudence Committee, to join with him because of the nature of his committee. There were also some joint sessions between the Senate Select Committee on Redistricting (“Senate Redistricting Committee” or “SRC”) and the joint House committee. TrA1065 (Hunter). Hearings were held on June 21, 2010 (San Antonio); July 19, 2010 (McAllen); July 20, 2010 (Laredo); July 21, 2010 (Corpus Christi); August 16, 2010 (El Paso); August 18, 2010 (Lubbock); September 20, 2010 (Dallas) (joint hearing of House and Senate committees); September 21, 2010 (Arlington); Septem- ber 22, 2010 (Richardson/UT-Dallas); October 4, 2010 (Amarillo); October 5, 2010 (Midland); October 18, 2010 (Beaumont); October 20, 2010 (Marshall); October 21, 2010 (Edinburg); October 27, 2010 (Abilene); Novem- ber 4 (San Antonio); November 17, 2010 (Austin); and November 20, 2010 (Houston) (joint hearing). 37. The Legislature did not make it a priority for House or Senate Redistricting Committee members to attend the field hearings. Committee members were not required to attend, and at least some were not consulted about their availability before hearings were scheduled. TrJ636, TrJ639 (Herrero); TrA1089 (Hunter). However, several members of the HRC attended many of the hearings and also served on the 18a HRC in the 82nd Legislature, including Reps. Peña, Alvarado, Harless, Hilderbran, and Veasey. Nine members of the 81st Legislature’s HRC were re-elected and served on the HRC in the 82nd Legislature: Villarreal, Peña, Alvarado, Eissler, Harless, Hilderbran, Keffer, Pickett, and Veasey. Most of these members attended some of the field hearings. In addition, some members who were not on the HRC in the 81st Legislature but served on the HRC in the 82nd Legislature attended many of the field hearings, including Reps. Alonzo, Hunter, and Phillips. 38. Sen. Kel Seliger, who would chair the SRC in the 82nd Legislature, attended the hearings held with the SRC. Rep. Burt Solomons, who would chair the HRC in the 82nd Legislature, did not attend any of the hearings. Tr1556 (Solomons). 39. David Hanna of the TLC attended about a dozen of the interim 2010 hearings, including Austin, McAllen or Edinburg, Laredo, Midland, Amarillo, Dallas, Arlington, Houston, San Antonio, and Corpus Christi. TrA1555-56 (Hanna). During the 2011 redistricting cycle, Hanna was Senior Legislative Counsel and provided legal advice to members and mapdrawers on redistricting issues. TrJ1145-47 (Hanna). In that role, Hanna gave presentations for members relating to some issues they might face, such as the County Line Rule and the VRA, and also authored a resource book titled State and Federal Law Governing Redistricting in Texas, which was available to members in March 2011 (except for the chapter on § 5 of the VRA). TrJ1148 (Hanna); US-357. 40. The 2010 Census data was not yet available at the time of the interim hearings, and no proposed maps were available. TrA1089 (Hunter); TrJ644 (Herrero). The main goal of the hearings was to let the public and 19a legislators know that redistricting would be occurring in 2011, and to give people an opportunity to speak. TrA1065, TrA1100 (Hunter). Chambers of commerce, LULAC, NAACP, various business organizations, the general public, and experts appeared and spoke. TrA1066 (Hunter). Members of both political parties and local officials and congresspersons attended. Id. However, the lack of census data or proposed maps rendered the hearings less useful in terms of substance and obtaining meaningful feedback from the public. TrA906 (Dukes). 41. The House does not use court reporters, so it did not make stenographic transcripts of the field hearings. TrA1068 (Hunter). However, audio/visual record- ings were made and were posted on the House website. Id. They were available for review by members of the public and the Legislature, and are still available in the House archives at http://www.house.state.tx.us. Id. Transcripts were also made at the request of the Texas Office of the Attorney General (“OAG”), but there is no indication that these transcripts were made available to the public, to members of the Legislature, or to the HRC during redistricting. TrJ642-43 (Herrero); TrA1091-92 (Hunter). In addition, there is no evidence that members of the 2011 HRC who did not attend the field hearings reviewed the testimony given at the 2010 field hearings. 42. The 81st Legislature HRC provided an interim report to the 82nd Legislature. US-340. It stated that the full HRC and subcommittees had conducted 17 interim public hearings throughout Texas to obtain public input as well as input from invited experts in the field of the federal census and Texas demographics. The interim report lists the dates and locations of the hearings; it does not include substantive 20a summaries of testimony or transcripts of the hearings. Id.; TrJ635-37, TrJ640-41 (Herrero). None of the testimony was made part of the legislative record. TrJ644 (Herrero). Rep. Carol Alvarado included a supplemental letter summarizing briefly some areas high- lighted by the public in certain geographic areas. US- 340. This letter stated that the Dallas community was very vocal about having a congressional district that could elect a Latino; the Austin community desired representation in a single congressional district; and the Houston community wanted a district to represent the burgeoning Latino community. Id. Committees do not always do a full writeup of their findings, and it would have been up to Chairman Jones how exten- sively to write up his findings. TrA1070 (Hunter). Rep. Herrero was asked to sign the report, but declined because none of the witnesses’ testimony was made part of the record. TrJ641, TrJ644 (Herrero). The report omits his signature line from the list of HRC members. 43. Although the interim field hearings may have succeeded in educating the public somewhat about upcoming redistricting, they were of limited useful- ness in terms of obtaining meaningful public input for legislators, and there is little indication that the 82nd Legislature or the mapdrawers paid much attention to the public testimony received at these hearings. 44. The SRC held a hearing on Wednesday, September 1, 2010 at the Capitol Extension in Austin, Texas to consider organizational matters and to hear invited testimony only. D-117 at 20; D-578. (A) Seliger noted that Doug Davis would be the Committee Director. Committee rules were adopted, and Hanna and Dyer of the TLC and Dr. Lloyd Potter, State Demographer, 21a testified. D-117 at 20. No public testimony was allowed. Id. at 22. (B) Potter noted that the most recent ACS data was from 2009, but that he also had 2010 projections that he could provide to the Senators. D-578 at 27. He noted that His- panics had seen the most growth and had contributed significantly to the population growth and that population growth overall was high in the suburban counties surrounding major urban areas. Id. at 22, 28. He also stated that there were 49 counties that were more than 50% Hispanic in 2009, compared to 34 in 2000. Id. at 23. (C) Dyer testified that TLC had been collecting election data by precinct for the last decade and had “primaries, runoffs, and general election results.” Id. at part 2, page 3. She said, “[A]ll of our data is available on the FTP site, available to the public, also, currently, and the new data will be available when it’s here, so we have a geographic area with current plans, precincts, VTDs, school districts, census geography, and ZIP codes by district, we have election returns, registered voters, [SSRV, . . .] and turnout from 1998 through 2008 right now, and we’ll have the 2010 primary up in September.” Id. at 12. (D) Hanna told them that for House and Senate maps, they could have a total 10% population deviation across districts without having to justify it. Id. at 13. However, he did call their attention to Larios v. Cox, 300 F. Supp. 2d 1320 (N.D. Ga. 2004) and noted 22a that it might change things. Id. at 14. Hanna said that to guard against a Larios problem, they could minimize deviation or be sure that deviations were not consistently disadvantaging one group. Id. at 16. He also pointed out the change for CVAP data from the decennial census to the ACS, noting that the margins of error were now larger, especially for smaller areas. Id. at 21. He stated, “to summarize on this CVAP, we need it for compliance with Section 2. We’re uncertain of the reliability of the new data, but it appears we’ll be able to use it in compliance with Section 2, for legislative districts, and we’re unclear when the data is gonna become available but, it looks like it may arrive late in the cycle the Legislature has available for conducting legislative redistricting.” Id. at 23. 45. In September 2010, Rep. Todd Hunter, Rep. Aaron Peña, Texas House Speaker Joe Straus’s legislative director Lisa Kaufman, Tom Phillips of Baker Botts, and a representative from the Texas OAG’s office went to Washington, D.C. and met with members of the Texas congressional delegation. They met with Congresspersons Eddie Bernice Johnson, Lamar Smith, Henry Cuellar, Charlie Gonzalez, Lloyd Dog- gett, Ron Paul, Sheila Jackson Lee, and others. TrA683 (Johnson); TrA1079 (Hunter). The purpose of the meeting was to make the congressional representatives aware that redistricting would be in the next year, let them know about the field hearings, and give them an opportunity to provide any input to the Redistricting and Judiciary Committees. TrA1079 (Hunter). Congressman Smith was designated as the coordinator for the congressional delegation’s map 23a drawing. TrA683 (Johnson). Smith was chosen as the liaison to the Texas Legislature because of his relationship with Straus. TrJ1475 (Interiano). Eric Opiela, who had worked on Straus’s re-election campaign in the fall of 2010, was designated as the attorney for the delegation and the person drawing maps for the delegation. TrJ1478-1479, TrJ1615 (Interiano); TrA683 (Johnson). 46. In October 2010, Speaker Straus hired Gerardo Interiano as his counsel for redistricting. TrJ1474, TrJ1615 (Interiano). Interiano is a Hispanic Republi- can and an attorney, and he had previously worked as a campaign manager, district director, legislative director, and deputy chief of staff for Congressman Smith. TrJ1474, TrA324 (Interiano), Interiano 8-9-11 depo. at 52. He had also worked on Straus’s primary campaign. TrA324 (Interiano). Interiano was hired as counsel to Straus and his agents to aid in redistricting, and he reported to the Speaker’s office. Interiano 8-2- 11 depo. at 22; TrJ1572, TrJ1923 (Interiano). His role was to advise the Speaker through the redistricting process (including on compliance with the VRA) and to draw maps. Tr1418, 1441 (Interiano); Interiano 8-2-11 depo. at 47. Interiano reported to Straus but also worked at the direction of Chairman Solomons. TrJ1923 (Bruce); D-113 at S101. 47. Interiano was trained on RedAppl by the TLC when he first became involved in redistricting. TrA335 (Interiano). RedAppl is short for Redistricting Appli- cation, and it is the TLC’s application developed inter- nally for use by the legislature and sponsored and public clients for redistricting purposes. TrJ227 (Dyer). Every legislator has access to RedAppl, and public access terminals were available in the TLC offices. 24a TrJ227-31 (Dyer). Interiano said that he spent approximately 1000 hours familiarizing himself with the RedAppl software after he was hired by Straus. TrJ1476. However, he testified at trial that he was not aware at the time he was mapping of how certain data was affected when precincts were split. TrJ1590. He explained, “I know for political data, which includes SSVR, when you split at the precinct level, it assumes that it’s a homogenous distribution of that political data information in every single block. So, for example, if . . . the precinct is 60 percent Abbott, it’s going to assume that every single block in that precinct is 60 percent Abbott as well. The same applies for SSVR. And I wrongfully assumed that that applied across the board through all of the metrics.” TrJ1590-91. He also attended the National Conference of State Legisla- tures, read articles, and did “self study” on redistricting law. TrJ1475-76 (Interiano). Before going to work for Straus, Interiano helped Congressman Smith organize the congressional delegation to work with the Legislature on congressional redistricting. TrJ1475. 48. Interiano was the principal drafter of the House plan. Tr1418, TrJ1472, TrJ1575 (Interiano). He also played a smaller role in drafting the congressional plan. Tr1418, TrA296-97 (Interiano). 49. Interiano and Opiela had known each other since approximately 2001 and were friends. TrJ1478 (Interiano); TrJ2085 (Downton). Opiela frequently went by the redistricting office, and he and Interiano had numerous contacts during the 2011 redistricting process TrJ1968 (Bruce); Interiano 8-9-11 depo. at 53; TrA317 (Interiano); TrA1727 (Downton). The redistricting office was a shared office between the Speaker and the HRC. TrA315 (Interiano). 25a 50. The 2010 general election was a good election for Republicans. TrA548 (Flores); Engstrom Corr. Rebuttal Report (docket no. 307-1) at 27; Joint Expert Ex. E-2 (Kousser report) at 73. The 2010 election featured reduced voter turnout among Hispanic and African-American voters compared to 2008, and older Anglo voters turned out at higher levels, nearing participation rates typically associated with a presidential election. Tr878-79 (Murray). As a result, the overall composition of the electorate was quite different than in 2008, and several districts that had elected Hispanic voters’ preferred candidate previously did not do so. Tr878-89 (Murray); Tr302 (Kousser); Joint Expert Ex. E-2 (Kousser report) at 73-76. Registered voter turnout in the 2010 general election was 38%. D- 661; TrA548 (Flores). Turnout in the 2008 presidential election was 59.5%. TrA549. 51. Francisco “Quico” Canseco, a Hispanic Republi- can, was elected to the U.S. House of Representatives to represent CD23, defeating incumbent Democrat Ciro Rodriguez. Canseco was not the Hispanic candidate of choice. Tr225, Tr302 (Kousser); Joint Expert Ex. E-2 (Kousser report) at 46. Canseco had previously run in 2004 and 2008, but lost in the Republican primaries. TrA565 (Canseco). In 2010, he was the only Hispanic in the Republican primary against several non-Hispanic candidates, and he narrowly won the runoff. D-35; PL-1031; TrA566 (Canseco); PL-1030 (he won the primary runoff by 722 votes). In the general election, Canseco received 49.39% of the vote (74,853 votes), and Rodriguez received 44.44% of the vote (67,348 votes). D-35. 52. Blake Farenthold, an Anglo Republican, was elected to the U.S. House of Representatives to represent CD27, narrowly defeating 27-year incumbent 26a Solomon Ortiz, a Hispanic Democrat. TrA226-28 (Kousser); Tr1870 (Alford). In the general election, Farenthold received 47.84% of the vote (51,001 votes) and Ortiz received 47.11% of the vote (50,226 votes). D-35. Ortiz was the Latino candidate of choice. Tr226 (Kousser); Joint Expert Ex. E-2 (Kousser report) at 47. 53. In Nueces County, Raul Torres (Hispanic, Republican) defeated incumbent Solomon Ortiz, Jr. (Hispanic, Democrat) to represent HD33 by a vote of 52.51% (12,499 votes) to 47.49% (11,306 votes). Connie Scott (Anglo, Republican), defeated incumbent Abel Herrero (Hispanic, Democrat) to represent HD34 by a vote of 53.96% (13,892 votes) to 46.04% (11,855 votes). Todd Hunter (Anglo, Republican) was re-elected without an opponent. 54. Jose Aliseda (Hispanic, Republican) defeated incumbent Democrat Yvonne Gonzalez Toureilles to represent HD35, by a vote of 52.81% (15,324 votes) to 47.19% (13,692 votes). John Garza (Hispanic, Republi- can) defeated incumbent Democrat David McQuade Leibowitz in HD117 in Bexar County, by a vote of 51.89% (14,705 votes) to 48.11% (13,635 votes). Dee Margo (Anglo, Republican) defeated incumbent His- panic Democrat Joe Moody in HD78 in El Paso by a vote of 52.41% (15,337 votes) to 47.59% (13,927 votes). Marva Beck (Anglo, Republican) defeated incumbent Democrat Jim Dunnam in HD57 in Waco/McLennan County. Aaron Peña, a Hispanic, was re-elected as a Democrat to represent HD40 in Hidalgo County. TrA88 (Peña). After the election, Peña announced that he was switching parties to become a Republican. TrA121-22 (Peña).

27a 55. On November 17, 2010, Opiela emailed Interiano (copying Lisa Kaufman of Straus’s office) with the subject “useful metric,” writing, Just had a thought I needed to get out before I forget it. The raw data to calculate this is going to be in the PL 94-171 dataset4 we’ll get in March (hopefully), but it would be really useful for someone to go in and calculate a ratio for every census block in the state of CVAP/Total Population, a ratio of Hispanic CVAP/Total Hispanic Population, a ratio of Spanish Surname RV/Hispanic CVAP, and a ratio of Spanish Surname RV/Total Hispanic Population (these last two have to be calcu- lated with the voter file overlaid with census data). It also would be good to calculate a Spanish Surname Turnout/Total Turnout ratio for the 2006-2010 General Elections for all VTDs (I already have the data for this for 2006-2008 in a spreadsheet, just need to gather it for every VTD for 2010). These metrics would be useful in identifying a ‘nudge factor’ by which one can analyze which census blocks, when added to a particular district (especially 50+1 minority majority districts) help pull the district’s Total Hispanic Pop and Hispanic CVAPs up to majority status, but leave the Spanish Surname RV [registered voters] and TO [turnout] the lowest. This is

4 Under 13 U.S.C. § 141, commonly referred to as “Public Law 94-171,” the Secretary of Commerce was required, by April 2, 2011, to complete, report, and transmit to each state the detailed tabulations of population for specific geographic areas within each state. States ordinarily use the P.L. 94-171 data to redraw districts. 28a especially valuable in shoring up Canseco and Farenthold. US-755; PL-1617; Quesada-253; NAACP-651; TrJ1536- 37 (Interiano). Opiela’s proposal is referred to in this litigation as the “nudge factor.” On November 19, Interiano responded, “I will gladly help with this Eric, but you’re going to have to explain to me in layman’s terms. Maybe you and I can sit down and go through this and you can show me exactly what you want next week or after Thanksgiving.” Opiela responded, “Happy to. Thanks Gerardo. Think of as ‘OHRVS’ Optimal Hispanic Republican Voting Strength. . . . a measure of how Hispanic, and Republican at the same time we can make a particular census block.”6 US-75. Although Interiano did not understand Opiela’s email initially, he eventually came to understand Opiela’s “nudge factor” proposal and how it could work. Interiano admitted that there was never any doubt in his mind that Opiela was suggesting drawing districts that would appear to be Latino opportunity districts because their demographic benchmarks were above a certain level but would elect a candidate who was not the Hispanic candidate of choice. TrA375-376 (Interiano) 56. On November 20, 2010, Opiela sent an email to Congressman Smith, which he also forwarded to

5 Defendants filed written objections to US-75 on the grounds of foundation, authentication, hearsay, and relevance and to Quesada-253 on the basis of hearsay. These objections are overruled. 6 “ORVS,” which stands for Optimal Republican Voting Strength, was a term used by some involved in redistricting to measure Republican voting strength in a proposed district based on an index of specific elections. D256_00002. Opiela has modified the term to Optimal Hispanic Republican Voting Strength. 29a Denise Davis, Straus’s chief of staff. US-767. In that email, Opiela stated that CVAP data was only an estimate and did not match up exactly with 2010 census data. He also advised that there was some uncertainty whether VAP or CVAP was the appropriate basis upon which to bring a § 2 vote dilution claim, and stated that “ultimately we need to run both measures on all maps we draw to know where we stand regardless of which way the courts go (CVAP or VAP).” Opiela further wrote, Gerardo [Interiano] and I were talking last night about the problems inherent in trying to protect both Farenthold and Canseco. It will be INCREDIBLY difficult to not have a packing claim and enhance Farenthold’s reelectibility. There is a ripple effect created by splitting Nueces from Cameron County in that the Cameron County district will have to go North to pick up Anglo voters if it doesn’t pick them up in Nueces, and there simply aren’t enough Anglo voters outside of Nueces in South Texas to pick up without reaching up to Wilson/Guadalupe/Eastern Bexar County. If we do this then we squeeze Cuellar up against the Border and pack him – unless we give Western Bexar to Cuellar to pick up Anglo voters there. However if we do this, then we severely damage Canseco’s reelect- ability. We could move Canseco up to pick up Anglo voters in Midland, but then we pair him with Conaway – which is unacceptable, and probably will lead to a Section 2 vote dilution claim on the basis of (the new) CD23. You

7 The hearsay objection (TrJ1479-80) is overruled because the exhibit is not offered or used for its truth. 30a see the problem? I think we should focus on CD23 for enhancement since it is the easiest to reconfigure without the ripple effect – principally by taking more of your district in, e.g. your western counties (especially since due to growth, you’re ‘heavy’ in terms of population), and potentially some of CD 20 if we have to enhance Hispanic VAP to offset (problem here is that CD 20 has to grow in size – it’s been growing slower than the state average). . . there’s a lot more possibilities working on CD 23 first. We definitely don’t have to reconfigure to his detriment since we can play a lot with the large body of people in Bexar to ensure his minority majority status AND make it still Republican – I don’t know yet whether he was the minority community candidate of choice – suspect he wasn’t – but once we get the VTD level election data from Nov. 2 we will know the answer to that question. Interiano noted that the referenced conversation was before the census data came out, and primarily “it was an issue of having two Republicans that had been elected in heavily Hispanic districts and what that would mean as far as how it would be drawn and where the populations would have to come from.” TrA343-44. Interiano was working for the Speaker at the time. TrA344 (Interiano). Interiano testified that “[o]ne of the big constraints in the case of Congress- man Farenthold was the role that Nueces County played and whether Nueces County would be the anchor of the district, whether it should go north, whether it should go south. We had had some of those interim hearings before redistricting began. And before we had the census data, one of the testimonies that I recall 31a from there was that Nueces County wanted to be the anchor of a district. And so we were trying to figure out how do we make all of those things work. And in the case of Congressman Canseco, where do you pull those populations from for those south Texas districts.” TrA344. Interiano also testified that he was aware that Opiela had a large trove of data available to him for drawing plans, but stated that he never saw the database or his computer. TrJ1493, TrA298. 57. On December 7, 2010, Interiano emailed Sam Davenport and Dyer of the TLC, asking for the information that Opiela had wanted for his “nudge factor.” D-262. Interiano cut and pasted the various ratios mentioned by Opiela in his email into the new email, and thus asked if the following information was available: “By census block, the following ratios: -CVAP/ Total Population; - Hispanic CVAP/Total Hispanic Population; -Spanish Surname RV/Hispanic CV; - Spanish Surname RV/Total Hispanic Population. By VTD, the following ratios: -Spanish Surname Turnout/ Total Turnout for the 2006-2010.” US-81; TrJ1482 (Interiano). This data was not available in an existing report, and was a custom data request. TrJ257, TrJ293-94 (Dyer). 58. Dyer responded to Interiano on December 8, explaining that TLC received the 2000 citizenship data at the block group level in 2002 but since that data was an estimate, it was not allocated down to the block level. D-262. She explained that the Census Bureau would release the 5-year 2005-2009 ACS data in December, which would have data down to the census tract level, but that it would be provided on 2000 census tracts rather than 2010 census tracts. She stated that they did “not expect to receive the 2010 census population data until late February or March, 32a which will include the total and Hispanic population,” and that they expected to receive the ACS estimates at the block group level during that same time period. Therefore, she explained, the answers to the questions were as follows: “Q 1. CVAP/Total Population. A. We can provide this information for 2000 by census block group, but not by census block. We do not have enough data yet to know what we will be able to provide for 2010, but we are working on this. Q 2. Hispanic CVAP/Total Hispanic Population and Q 3. Spanish Surname RV/Hispanic CVAP. A. Same as answer to 1. Q 4. Spanish Surname RV/Total Hispanic Population A. We can provide estimated Spanish surname registered voters by voting tabulation district (VTD) or by block for 2000 and total Hispanic population for 2000 by block. (VTDs are census precinct equivalents that are composed of whole census blocks.) By VTD, the following ratios: Q5. Spanish Surname Turnout/total Turnout for the 2006-2010 [sic] A. We can provide Spanish surname registered voters by VTD for 2006- 2008 (primary and general) and 2010 primary. We expect to have the 2010 general election data ready in late March after we receive the 2010 census population. It is part of the 2010 general election data base, which we are in the process of collecting from the counties and entering into our database. We can provide total turnout by VTD or allocated to the block level for 2006-2008 (primary and general) and 2010 primary. We do not have Spanish surname turnout. You may be able to use voter history files collected from the individual counties to calculate the Spanish surname turnout, but we do not have it.” 59. Using the information in RedAppl at the time, one could identify VTDs that were relatively high in Hispanic population but relatively low in total turnout (but not SSVR turnout). TrJ265-67 (Dyer). In addition, 33a Spanish Surname voter turnout data was available from the Texas Secretary of State, and had been since 2006 or 2008. TrJ299, TrA749 (Dyer). This data included a voter’s name and address, whether they had a Spanish surname, their precinct, and whether they voted. TrA749-51 (Dyer). 60. Interiano forwarded Dyer’s response email to his personal email account, and then forwarded the email from his personal email account to Opiela, Hull, and Kaufman on December 8. US-82; US-83.8 Interiano responded to Dyer’s email on December 9, “If you could provide me whatever data is already available from the requests below, it would be helpful.” D- 262. Dyer responded, “The only requested item we can actually do at the block level is Spanish Surname VR / Total Hispanic Population. We will generate that and send it. For which election (2008 general?) would you like the SSVR data?” Id. Interiano forwarded this email to Opiela asking, “See below . . . what do you want?” Id. Opiela responded, “08 and 10 would be nice. 08 is probably the only practical data.” US-85. Interiano wrote back to Dyer on December 10, “2008 is fine. And is that data going to be to the block level?” Id. Davenport responded, “Yes, the data will be at the block level. I’ll try to build the file for you today. It may be Monday.” Id. Interiano then replied to Opiela, “They won’t have 10 data yet. So I went ahead and asked for it for 08. We’ll have it on Monday.” Quesada-150. 61. The file was sent on Monday December 13, 2010. Davenport emailed Interiano with the subject “Census Data” and wrote, “Attached is the block level

8 The hearsay objections to US-82, US-83, US-85, and Quesada-150 are overruled because they are not admitted or considered for their truth. 34a file which contains the following fields: [list of fields].” US-185. Interiano forwarded the email to his personal email account. Interiano then forwarded the data he received from TLC to Opiela (and others). TrJ1485 (Interiano). The data that TLC sent Interiano would enable one to identify census blocks that had a low turnout in the 2008 election and census blocks with relatively high SSVR rates. Tr59 (Dyer). 62. On or about December 21, 2010, the U.S. Secretary of Commerce reported to the President the tabulation of population for each of the fifty states, as determined in the 2010 decennial census. Given the increase in population in Texas between 2000 and 2010, the number of seats in the United States House of Representatives for Texas increased from 32 to 36 seats. Tr903 (Downton). The ideal population for each of the 36 congressional districts in Texas is 698,488. The Texas House of Representatives has 150 members, each elected in a single-member district. Based on the 2010 U.S. Census, the ideal population size for each House district is 167,637. 63. ACS data is the only data source for which citizenship information is available in small geographic areas. Tr182 (Chapa). Because ACS data is based on a sample, it is subject to sampling error. CVAP data is fairly accurate at the statewide level; the margins of error at the state level are relatively small. Tr1680 (Rives). Rives testified that HCVAP data for a congressional district population of 698,000 should have a relatively small margin of error and estimates should be well within the range that the Bureau considers statistically reliable. Tr1680. He stated that he would feel somewhat less confident at 167,000, the size of a House district, but that population is above the 35a 65,000 threshold. Tr1680-81. He would not feel comfortable with the accuracy at the precinct level, but he was not aware of any current HCVAP estimates that were more accurate. Tr1681-82. 64. For redistricting, the data used is a 5-year average because it provides enough data for precision down to the census tract or block level. Tr1100 (Ansolabahere). The TLC used a Special Tabulation of the 2005-2009 ACS requested by DOJ, which produced CVAP by race and Hispanic origin for areas down to the block group (the smallest geographic unit for which ACS data can be published). Tr1670 (Rives). The figures contained in the DOJ Special Tabulation are the basis for the CVAP estimates published by the TLC. Tr1670 (Rives); Joint Expert Ex. E-1 (Chapa report) at 11. 2011 65. Minorities accounted for most of the CVAP growth from 2000 to 2010. TrA935 (Ansolabahere). 2011 was the first year ACS data (the 2005 to 2009 5- year ACS estimates and the DOJ Special Tabulation) were used for redistricting. ACS 5-year estimates are based on five preceding years, which means that the information is not current with the release date, and the 2005-2009 ACS data lagged behind the 2010 Census data. Tr1678 (Rives); Tr713 (Korbel); Tr1090, Tr1107 (Ansolabahere). The 2005-2009 ACS citizenship data/DOJ Special Tabulation used by the Texas Legislature to redistrict in 2011 underestimated minority CVAP in 2010/2011. Tr1032 (Murray); Joint Expert Ex. E-1 (Chapa report); Joint Expert Ex. E-15 (Ansolabahere report) at 12 (“The calculations reveal that the ACS slightly over estimates the White CVAP, and substantially underestimates the Black and His- panic CVAP.”). Dr. Ansolabahere testified that the total number of persons estimated by ACS is about 1.3 36a million fewer than the 2010 census (almost two congressional districts worth of people). Tr1101-02. The ACS overestimates (compared with the census enumeration) White Non-Hispanics by 30,000 persons, and underestimates Blacks by 200,000 persons and Hispanics by 900,000 persons. Joint Expert Ex. E-15 (Ansolabahere rebuttal report) at 7. Therefore, His- panics and Blacks account for 1.1 million persons of the 1.3 million persons in Texas underestimated by the ACS. Id. 66. Although the Census Bureau cautions against viewing a 5-year estimate as reflecting a midpoint, the experts testified they would be inclined to view a 5-year estimate as reflecting the midpoint year, and that the 2005-2009 ACS data likely reflected 2007 citizenship/population. Tr1678 (Rives); TrJ890, TrA795 (Fairfax); Joint Expert Ex. E-15 (Ansolabahere 8-30- 11 Rebuttal Report) at 5-7 (one may interpret the 2005-2009 ACS as an estimate of the population in 2007); see also TrA933 (Ansolabahere) (noting that the 2008-2012 ACS has a midpoint of 2010 and it aligns very closely with the 2010 census enumeration). Dr. Fairfax performed a verification test that supports his conclusion that the 2005-2009 ACS data reflected demographics in 2007. TrJ890-91 (Fairfax). 67. From 2005 to 2009, the number and proportion of the HCVAP increased. Tr180 (Chapa). The share of HCVAP of Texas in 2000 according to the 2000 Census data was 22.4%. According to ACS data, Hispanics accounted for 24.6% of HCVAP in 2005, 24.6% in 2006, 24.7% in 2007, 25.2% in 2008, and 25.5% of HCVAP in 2009. Joint Expert Ex. E-1 (Chapa Table 7). The 2009 ACS 1-year estimate of CVAP of 3,944,088 (25.5%) is higher than the estimate from the 2005-2009 DOJ special tabulation, which is 3,674,800 (24.7%). Tr180 37a (Chapa) (referring to Table 7 in his report); Joint Expert Ex. E-1 (Chapa report) at 11 (noting that the DOJ Special Tabulation substantially underestimates the HCVAP number and proportion in 2009). 68. According to the 2006-2010 ACS, the State of Texas has a CVAP of 15,276,966 (±14,248) persons, including 8,800,442 (±5,099) non-Hispanic white persons (57.6%), 3,889,571 (±11,549) Hispanic persons (25.5%), 1,938,918 (±4,208) non-Hispanic black persons (12.7%), and 419,716 (±3,618) non-Hispanic Asian persons (2.7%). Docket no. 1085, Ex. 5 (judicially noticed per text order 6/20/14). 69. Dr. Chapa opined that the ACS data and DOJ Special Tabulation should undergo additional analysis to be useful and reliable for redistricting. Dr. Chapa “updated” the 2005-2009 ACS estimates with 2010 population counts to provide conservative 2010 HCVAP estimates for districts in various House and congressional maps at issue in this litigation. Joint Expert Ex. E-1 (Chapa report) at 12. Overall, these estimates are slightly higher than the TLC data, though not in all cases. Joint Expert Ex. E-1 at 14-15. Chapa estimated a 2010 statewide HCVAP of 26%. Id. (Table 8). The ACS 1-year estimate for 2010 shows total CVAP of 15,854,093, including 8,952,806 (56.5%) non-Hispanic White persons, 4,180,024 (26.4%) Hispanic persons, 2,048,450 (12.9%) Black persons, and 463,558 (2.9%) Asian persons. Using the midpoint of the 2008-2012 ACS data and the census data, Dr. Ansolabahere estimated the 2010 statewide HCVAP at 26.5% (an increase of 1.24 million persons from 2000 to 2010), BCVAP at 13% (an increase of roughly 430,000 persons), and Anglo CVAP at 56.4% (an increase of 660,000). TrA934 (Ansolabahere); Rod-912 (report) at 9. 38a 70. In the 2011 session, the Texas Legislature had to pass redistricting maps for the Texas House, Texas Senate, Texas State Board of Education (“SBOE”), and the U.S. House of Representatives. The Legislature also had other bills, such as the budget, sanctuary cities, voter ID, and the sonogram bill (those four were declared emergency items by the Governor). There were also a number of agency bills that required sunset review. Tr108 (Martinez Fischer). 71. Interiano and others went to D.C. in early 2011 to meet with many of the congressional offices, both Republican and Democrat. PL-311 (meeting with Flores, Farenthold, Smith, Canseco, Poe, Hensarling, Hinojosa, Carter); Interiano 8-2-11 depo. at 77; Interiano 8-9-11 depo. at 45; Interiano 8-26-11 depo. at 69-70. 72. The 82nd Regular Legislative Session began on January 11, 2011. The Texas House of Representa- tives consisted of 49 Democrats and 101 Republicans. Tr996 (Downton). Eight of the 101 Republicans were minorities (two African-American, five Hispanic, and one Asian) and the rest (93) were Anglo. Most Demo- crat members of the House were minority. Tr1630 (Solomons). Of the 49 Democrats, 8 were Anglo, 25 were Hispanic, 15 were African American, and one was Asian. The Texas Senate consisted of 19 Republi- cans and 12 Democrats. 73. Because CVAP data was not yet available at the start of the session, mapdrawers working on the proposed House map relied on SSVR data obtained by TLC from the Texas Secretary of State. The data was reported as non-suspense SSVR and total SSVR, which included both suspense and non-suspense SSVR. Sus- pense voters were those who had not responded to a residence confirmation notice or whose renewal registration certificate had been returned because the 39a addressee had moved, such that the voters’ registration was subject to cancellation. As a result, total SSVR was usually slightly different from non- suspense SSVR. Before 2011, TLC reports reflected non-suspense voter registration; but beginning in 2011, all TLC reports reflected total voter registration. PL-728_16. However, when modeling in RedAppl, only non-suspense SSVR was shown. SSVR was reported as an election statistic (not a demographic statistic) in RedAppl, meaning that it was not available for labeling or shading below the precinct level. TrJ1599 (Interiano). 74. Although SSVR is considered relatively accurate, there are some problems with its accuracy because some Hispanic people do not have Hispanic surnames, and some people who are not Hispanic do have His- panic surnames. PL-728_16. Dr. Ansolabahere testified that there are 10-15% more people identifying themselves as Hispanics than have a Hispanic surname. Tr1114. Dr. Rives and Dr. Murray also agreed that SSVR has accuracy issues. Tr1688 (Rives); Tr1038 (Murray). Dr. Murray also testified that SSVR will be lower than HCVAP, such that a 50% SSVR district will almost certainly be majority HCVAP. Tr1056. 75. On January 17, 2011, Interiano emailed Dyer to ask questions on behalf of Opiela concerning TLC political data indicating election returns (votes) Opiela had imported matching up with the geography, stating that he “got the question below from another office.” Dyer responded, and Interiano forwarded the answer to Opiela. US-94. 9 On January 20, 2011, Opiela

9 Defendants’ hearsay objection is overruled because this exhibit was not offered for its truth. 40a emailed Interiano asking if TLC could provide him a shapefile with 2010 VTDs. PL-311. 76. On January 24, 2011, Rep. Veasey (African- American, Democrat) offered an amendment to H.R. 4 that would establish rules and procedures for redistricting that comply with the VRA. Solomons (Anglo, Republican) moved to table the amendment, and it was tabled 93 to 47. PL-223. Rep. Dukes testified that when an amendment is tabled, it cannot be introduced again during the session. TrA900. 77. In February 2011, the TLC issued a booklet called “Data for 2011 Redistricting in Texas” that outlined the various data available. PL-728. It also noted the limitations of allocated data, including that allocation based on voting age population assumed that all blocks within a precinct exhibited the same voting pattern, which is not likely to be the case. PL- 728_0018. The population data in RedAppl comes from the Census Bureau, and the precinct/ VTD data and election returns, turnout, and SSVR data comes from election data collected from the counties for every election. TrJ228 (Dyer). In 2011, RedAppl contained the following information/data: election returns for races (primaries, runoffs, and general elections) from 2002 and 2010 (election returns from 1996 to 2010 were available through maps and reports that could be run by TLC separately); data on total voter registration; SSVR; voter turnout; selected county and city elections through the 2010 general election; total population and voting age population by race, including black, Hispanic, black and Hispanic, Anglo (white only), and other. TrJ228, TrJ234-35 (Dyer); PL-728. CVAP data was not available in RedAppl but was available by requesting reports from TLC. TrJ235 (Dyer). RedAppl contains the following geographic 41a building blocks: counties, cities, VTDs, tracts, block groups, and blocks. TrJ236 (Dyer). A VTD is the equivalent of a voting precinct but follows census geography. 78. RedAppl will display selected statistics, chosen by the user, in a panel at the bottom of the screen as the mapdrawer works on the map. TrJ237, TrJ275 (Dyer). Available statistics include SSVR and specific election results by percent. TrJ238 (Dyer). As districts are changed, including at the block level, the statistics change accordingly. TrJ241 (Dyer). Election data is allocated within the system down to the block level, but since no one knows how specific people voted, the election returns are allocated the same across the precinct/VTD. TrJ280 (Dyer). When precincts are split, RedAppl allocates the number of votes accordingly but still maintains the election return/ percentages the same across the portions. Id. Therefore, data is less accurate for split precincts, and the more split precincts there are in a district, the less accurate overall the data will be. TrJ280-81 (Dyer). 79. RedAppl permits the user to shade geographic areas by percent turnout. TrJ239 (Dyer). RedAppl also displays the number of votes cast in a particular VTD in a particular election. TrJ239-40 (Dyer). RedAppl has block level shading for HVAP. TrJ240 (Dyer). Someone working on a map in RedAppl would be able to view the concentration of Hispanic VAP and the number of votes cast in specific precincts. TrJ241 (Dyer). RedAppl lets a user shade for population data down to the block level, and for election data at the county and VTD levels. TrJ276 (Dyer). All shading and labeling data is available at the VTD level. Id. No political shading is available at the block level. TrJ279. No CVAP data is available by block. TrJ290 (Dyer). SSVR is an election statistic, not a population 42a statistic, so it is not available below the precinct level. TrJ277 (Dyer). At the block level, only population by race/ethnicity and VAP is available. Id. However, as noted, changes made to a district at the block level will change election statistics shown in the statistics bar. RedAppl also has a feature that allows a user to bring a plan into the Google satellite image, allowing a user to determine whether stadiums and other such features are in a district. TrJ256 (Dyer). 80. The Legislature had no express policy against splitting precincts. TrJ1592 (Interiano). Similarly, the TLC does not have an express policy against splitting VTDs and does not have any written guidance instructing individuals not to split VTDs. TrJ294 (Dyer); TrJ2020-21 (Downton); TrJ1203 (Hanna). However, splitting precincts is highly disruptive to communities and discourages voting because people do not know where to go to vote. TrJ138 (Arrington). It is especially disruptive for those people who find it most difficult to vote because of socioeconomic status. TrJ139 (Arrington). Splitting precincts is therefore disfavored. TrJ734 (Pickett); TrJ139 (Arrington); Tr249 (Seliger) (agreeing that not splitting precincts would be part of Texas’s traditional redistricting principles, but stating that it was “not strictly adhered to”). Downton admitted that the mapdrawers asked members not to split precincts. TrJ2078. There are race-neutral reasons for splitting precincts, including: ensuring population equality; following city boundaries and roads; including financial supporters, a member’s home, airports ,and government buildings; and complying with the VRA and the Texas Election Code (which requires precincts to have 100 to 5000 residents). TrJ177-78 (Arrington). 43a 81. On February 7, Interiano emailed Hanna with the subject “Quick Question.” US-153. He wrote, “Under Section 2, would a majority single minority district get preference over a majority mixed minority district? [I]n other words, let’s say that X creates a district that is 65% Hispanic, Y creates a district that is 45% His- panic and 15% black, and Z creates a district that is 51% Hispanic and 10% black. Under those circumstances, which one would pass muster with the Voting Rights Act, knowing that it’s POSSIBLE to create the 65% district. Hope that makes sense.” Hanna responded, “Which district performs for Hispanics?” Interiano replied, “Let’s say both the 51 and the 65.” Hanna wrote back, “If the 51 truly performed for h then it wouldn’t matter. The reality is it won’t and you’ll need a higher percentage. Once you reach performing levels your choice until you start packing. Or diluting the Black vote by splitting it up.” 82. Senator Kel Seliger (Anglo, Republican), who represented SD31 in the north panhandle, South Plains, and Permian Basin area, was the Chair of the SRC in 2011. TrA218-19. The members of the SRC in the 82nd Legislature were: Chair Kel Seliger, Vice Chair Mario Gallegos, Jr., John Corona, Kevin Eltife, Craig Estes, Tony Fraser, Juan Hinojosa, Joan Huffman, Eddie Lucio, Jr., Dan Patrick, Carlos Uresti, Jeff Wentworth, Royce West, Tommy Williams, and Judith Zaffirini. PL-221 Seliger is not an attorney, and he had no redistricting experience at all. TrA262 (Seliger). Seliger hired Doug Davis, who had worked on redistricting in 2003, to be the Committee Director. TrA280 (Seliger). The SRC also hired outside counsel to advise it—David Guinn and Michael Morrison from Baylor University, and Bob Heath from Austin. Id. 44a 83. The HRC was formed early in February. On February 9, 2011, Burt Solomons (Anglo, Republican), who represented HD65 in Denton County, learned he was to be the Chair. PL-311; TrJ1067, TrJ1928-30 (Bruce). Hispanic legislators on the HRC included Mike Villarreal (D) (vice-chair), Aaron Peña (R), Carol Alvarado (D), and Roberto Alonzo (D). Tr110 (Martinez Fischer). Other members of the HRC included: Jimmy Don Aycock (R), Dan Branch (R), Ron Eissler (R), Charlie Geren (R), Patricia Harless (R), Harvey Hilderbran (R), Todd Hunter (R), Jim Keffer (R), Jerry Madden (R), Larry Phillips (R), Joe Pickett (D), and Marc Veasey (D). D-157. Veasey was the only African- American on the HRC. Tr795 (Turner). Therefore, the HRC consisted of 12 Republicans (11 Anglo, 1 His- panic) and 5 Democrats (1 Anglo, 3 Hispanic, and 1 African-American). 84. Congressman Lamar Smith stopped by Solomons’ office the day his chairmanship was announced and introduced Eric Opiela. TrJ1968 (Bruce). Soon after, the HRC hired Ryan Downton as its general counsel and primary map drawer. Tr903 (Downton); TrJ1928, TrJ1956 (Bruce). His role was to assist in drawing maps and to provide legal advice to the HRC and Solomons. Tr903 (Downton). Downton’s prior redistricting experience consisted of “self-study.” Downton 8-12-11 depo. at 10-11. Downton was lead staffer in charge of and principal drafter of the congressional plan. Tr903 (Downton); Tr1418 (Interiano). Downton also drafted or worked on portions of the House map, including Harris County, Dallas County, Tarrant County, Hidalgo County, and El Paso County. TrJ1989, TrJ2016, TrJ2025-29 (Downton); Downton 8-31-11 depo. at 73. Downton met Opiela early in the process. TrJ2085 (Downton). Opiela, Interiano, and Downton 45a had frequent conversations and contact throughout the redistricting process. 85. HRC resources included the TLC, Speaker staff (Interiano), and Baker Botts. TrA1063 (Hunter); TrJ1025 (Solomons). However, the availability of Baker Botts attorneys was not shared with minority legislators or minority members of the HRC. TrJ1958 (Bruce); TrJ1233, TrJ1288 (Thompson). Rep. Aycock, an Anglo Republican member of the HRC, also testified that he was not told about Baker Botts. TrJ1754. Solomons intended that other legislators would come to him and his staff, and he would then ask the OAG or Baker Botts for advice or answers and “try to get back with them.” TrJ1026 (Solomons). Solomons met with staff from the Texas OAG and they offered assistance with redistricting. TrJ1013, 1068 (Solomons). The OAG conducted racially polarized voting analyses (“RPVAs”). Solomons and redistricting staff also received from the OAG regression analysis reports on proposed districts in redistricting plans. TrJ1015 (Solomons). House redistricting staff relied on those reports and used them to identify the minority candidate of choice and whether they would prevail in proposed districts. TrJ1015-16 (Solomons). 86. Bonnie Bruce was Solomons’ chief of staff and the HRC Committee Clerk. TrJ1017 (Solomons); TrJ1920-21 (Bruce). Bruce was the managerial super- visor for the HRC; everyone reported to Solomons. TrJ1923 (Bruce). Bruce, Downton, and Interiano shared a suite of offices at the Capitol. The HRC and Speaker Straus had offices there and shared a conference room. TrJ1923 (Bruce); TrA1727-28 (Downton). The redistricting office was a shared office between the Speaker and the redistricting committee, and it was Interiano’s and Downtown’s primary office. TrA315. 46a 87. Each member of the Legislature is given a RedAppl account, which may be used by the member or the member’s designated staff member(s). TrJ229 (Dyer). Members would contact Dyer’s section for help with RedAppl and for special reports. Id. RedAppl accounts generally use the first four letters of the last name as the identifier. TrJ230-31 (Dyer). The strj account was assigned to Speaker Joe Straus. TrJ232 (Dyer). Interiano was the primary user of this account. TrJ244, TrJ1497 (Interiano). The hrc1 account was assigned to Solomons as chair of the HRC. TrJ232 (Dyer). Downton was the primary user of this account. TrJ1008 (Solomons); TrJ1066; TrJ1990 (Downton). The solo account was assigned to Chairman Solomons, and Bruce was the primary user. TrJ1008 (Solomons). A legislator can only log into their account on a dedicated computer in their office or on their laptop, after logging into the computer/network. TrJ1971 (Bruce). 88. Speaker Straus told Solomons to pass a House map during the session; they wanted to avoid having the LRB draw the map. TrJ1926 (Bruce). Solomons did not know redistricting law or have experience on VRA issues, though he knew the maps had to comply with the VRA. TrJ1028-31, TrJ1068, Tr1554, Tr1569, Tr1583 (Solomons) (“I am not experienced or knowledgeable about redistricting law.”). Solomons relied on the TLC, the OAG, and his staff to determine whether the maps met legal requirements. TrJ1030-31, Tr1591 (Solomons). Solomons agreed to take the Chair position because he was assured that he would have “plenty of people working that could do all of the detailed work.” TrJ1072 (Solomons). Solomons did have experience passing complex, controversial legislation, and he believed his role was to shepherd the legislation. TrJ1925-26 (Bruce); TrJ1072, Tr1634 (Solomons). 47a Solomons delegated responsibility and maintained general oversight. TrJ1073 (Solomons). Solomons said his goal was to pass a House map that would hold up and to make the map a member-driven map. TrJ1069 (Solomons). Solomons asked members to submit proposals to the committee, to work with committee staff, and to seek guidance from TLC if necessary. TrJ1071 (Solomons). 89. On February 16, 2011, the SRC held a public hearing to solicit public testimony and to hear invited testimony from Dyer and Hanna of the TLC. D-117 at 33; D-589 (transcript). This hearing focused mainly on the Senate map. A number of witnesses testified that the Travis County minority community worked in coalition and should be kept together in a district. Dyer informed the Committee that the TLC would be receiving the detailed population census data (the P.L. 94-171 data) on February 17. She stated that RedAppl would be ready within a week and that the “Data for 2011 Redistricting” booklet had been published. D-589 at 7. Seliger remarked that David Hanna had a tremendous amount of experience in redistricting and ventured to say that “any configuration in a map that is going to be seriously considered by the Committee or by the Legislature, is going to be scrutinized by Mr. Hanna.” Id. at 7-8. Hanna informed the Committee that the 2011 edition of Texas Redistricting Law had been released, except for the chapter on § 5 preclearance because they were awaiting final rules from the DOJ. He also stated that they had been working on the citizenship data and the initial tests indicated they would “have some useful citizenship data for purposes of Section 2 of the Voting Rights Act” and that they hoped to have it in three or four weeks. Id. at 8. 48a 90. On February 17, 2011, the Census Bureau released the P.L. 94-171 data from the 2010 Census to the State of Texas, which provided the first look at population counts for small areas and race, Hispanic origin, voting age, and housing unit data from the 2010 Census. D-129. 91. On February 17, Hanna emailed Denise Davis with the subject “redist issues.” D-192; US-102. He wrote, “1. Dallas Lose two seats. Both will have to be R’s b/c all D seats are minority. It may get worse. If it is possible to draw third Hispanic seat, you’ll need to do that too. That would be a 3rd seat. Four Black seats (as now) looks doable. 2. Houston- Like last time Houston comes out to 24.41 seats. Many will want to round up again as LRB did but this caused legal issues w. county line rule. Politically popular w. Harris County but legally more risky. Should do 24 but this will mean the loss of another R seat since all D seats are minority. (Hochberg has Hispanic seat). 3. Corpus - Two seats only; three R’s. And worse one of the seats will probably have to be more Hispanic than the other and probably elect a D. Not sure on this but preclearance likely an issue here. 4. Beaumont - Losing people and Black seat real short too. This will be tough to solve. Pop loss in W Texas, though El Paso looks OK for five seats. Pop loss in East Texas and limited number of valid county combos.” Hanna testified that he first determined, applying the County Line Rule, how many seats each of the whole counties would get. TrJ1185. Hanna said he made a cursory review and made a mistake on the Hochberg seat (HD137 in Harris County) being a minority seat because it has an HVAP in the 60s. Later he found the SSVR was in the 20s and he was not aware of a theory where that would be a protected seat, so his assessment changed. Id. 49a 92. On February 18, Denise Davis forwarded Hanna’s email to Interiano. D-132; US-102; Quesada- 251. Interiano responded to Davis stating, “[Hanna] and I went through all of this yesterday afternoon and through some of the first things that we need to look at as soon as RedAppl is up and running. I also visited with Wayne Smith [Rep. from Houston] and Solomons and mentioned the issue about Houston and immediately both of them leaned towards keeping Harris County at 25 seats. The big issue with Corpus is that two of the current seats are 65% Hispanic (Scott, Torres), but the county as a whole is only 60%. Mean- ing that all of the Anglos live in Hunters [sic] district, so it really is going to be a tricky issue. This might be one of the things to mention to Nina Perales when we meet next week and see if she has any thoughts on whether there should be two 60% seats or one 65% and one 55%. As far as Dallas goes, the ones that are going to be at risk are Sheets, R. Anderson, and Burkett, but Hartnett may also have issues because Branch has to pick up population as well and he told me last night that he would like to pick it up from Hartnett and give up some other portions to some of the minority seats. The issue with Beaumont is that currently there is a county split there for Deshotel [African-American, Democrat]. As soon as we get RedAppl running, we’re going to need to see the best way to get his percentage up, which may require us to keep that county split. I’m pretty convinced that this is where you have to start the map drawing in East Texas for various reasons but mainly because he is the only VRA district in East Texas. The big winners are clearly the suburbs. Tarrant, Denton, Collin, Fort Bend, Montgomery, Williamson, etc. won a great deal of population and that’s where the shifts are going to be. I also think that West Texas will likely lose 2 and East Texas will lose 50a 3. Just by adding everything that Dave mentioned below plus those losses in the East and West, I think you’re looking at a minimum loss of 8-9 Republican seats.” 93. On February 21, RedAppl was available with 2010 census population, 2002 through 2010 election data, and incumbent residence information. D-130. On February 21, Opiela requested a 720 report from TLC. US-730; TrA761 (Dyer). The 720 report is a block election data report, in which the election data is allocated to the block level. US-729. He had previously received a 720 report on January 28. US-730; TrA761 (Dyer). On February 21, Sam Davenport (TLC) emailed Interiano an attachment, “2010_hispanic_pop.pdf” and wrote, “Attached is the 2010 version.” US-186A. The attachment is a shaded map titled “2010 Hispanic Population by Block Group.” He had previously sent “Hispanic maps’ for 1990 and 2000. US-101. 94. On February 23, 2011, the HRC issued a notice of hearing on March 1 to take only invited testimony on the 2010 Census data, a notice of hearing on March 15 to take invited and public testimony on potential Texas SBOE districts in light of the Census data, and a notice of hearing on March 24-25 to take invited and public testimony on potential Texas House districts in light of the Census data, and a notice of hearing on April 7 on congressional redistricting. D-116 at 127, 131, 135, 141. 95. On February 24, 2011, Solomons went to the floor and discussed “drop-in counties” in the Texas House map. Certain counties were referred to as “drop-in counties” because it was possible to draw a whole number of districts within the county while staying within permissible population boundaries. Solomons explained that there were eight—Dallas, 51a Tarrant, Harris, Bexar, El Paso, Travis, Denton, and Nueces Counties—and asked the delegations of those counties to get together to get a consensus map for Solomons and his staff. TrJ1930-35 (Bruce). He wanted the representatives in the drop-in counties to work together and to have senior legislators guide the delegations. TrJ1074 (Solomons). Solomons was asked how many districts would be in Harris County, and he said 25. TrJ1931 (Bruce). Bruce said this caused the staff “a small panic attack” because that issue had not been decided, so they alerted him to this fact when he came off the floor, and he told them to clarify it. Id. Solomons’ office sent out a memo to members that same day clarifying that the decision about how many districts would be in the drop-in counties had not necessarily been made. Id.; D-72. The memo reiterated that notices had been posted for hearings in March and April and stated that “the Redistricting Commit- tee is actively seeking your assistance and input.” D- 72. It further stated, “The numbers I provided on the House floor relating to the number of districts that will be wholly contained within the larger urban counties represent a preliminary calculation of the number of ideal districts each county could receive. Only after the committee takes testimony will we be able to consider the actual number of districts that each county receives. The members will have to determine this number based on legal standards governing one-person, one-vote and the Voting Rights Act, the requirements of the Texas Constitution, as well as the policy choices of the members. If you need any guidance on the criteria that a district must meet under state and federal law, as well as preclearance I would encourage you to ask either the redistricting staff or if you need to ask confidential questions please contact the redistricting attorneys in the Texas Legislative Council.” D-72. 52a 96. On March 1, the HRC heard invited testimony from the TLC and Office of State Demographer on the 2010 Census data. D-116 at 127-28; D-4; TrJ1936 (Bruce); D-590 (transcript). There was no public testimony at this hearing. (A) There was no plan available yet. TrJ1937 (Bruce). Solomons announced Downton’s appointment as Public Information Officer for the HRC and requested that all public inquiries for information from the Commit- tee be referred to him. Jeff Archer and David Hanna of the TLC and Lloyd Potter from the Office of the State Demographer were the only witnesses. D-116 at 130; D-4. Potter noted that the counties surrounding the most populated urban counties had grown very dramatically, in particular Collin, Denton, Fort Bend, Rockwall, Fort Bend, and Montgomery Counties, and around Travis County. D-590 at 7, 9. Potter noted that his population breakdown by ethnicity, etc. was from 2000 to 2009 and he would not likely have the 2010 data until “probably . . . after you guys are done with your work.” Id. at 15. He also acknowledged there was undercount in the census, but it did not seem as bad as some had feared. Id. at 27. (B) Hanna and Archer gave testimony concerning redistricting law and requirements. Hanna stated that RedAppl training was available, as was the TLC’s legal book, except for the chapter on preclearance. Id. at 38. Hanna did a presentation on the County Line Rule. Id. at 38; D-124. Hanna 53a stated that there are two federal law exceptions to the County Line Rule—one person, one vote, and the VRA. He said, “We haven’t yet had an instance where the Voting Rights Act has caused us to cut a county line but I think it still remains out there as a hypothetical possibility.” Id. at 39. He further said that for the Texas House districts the law “allows for a ten percent deviation without justification . . . .” Id. at 40. But he noted that the case of Larios v. Cox might mean that “the old ten percent rule, which was a safe harbor, may not be any more.” Id. Hanna warned that “[i]f all [of a] certain kind of district are drawing small, then we start to run into the sort of issues that they had in Cox. Or if all the districts are drawing big, we start to run into that question. That’s not fully developed jurisprudence yet but that starts to get me a little bit concerned.” Id. at 45. (C) Hanna later stated, “we found out that there’s no way for Ellis County to be joined with another whole county. What this means is that you’re going to have to cut a county, and this is to comply with the One Person, One Vote Rule. The cut county could be either Ellis County itself or one of the counties around Ellis County, and the cut has to be related to making, you know, a district that fits within the One Person, One Vote situation. Looking at that also, I notice that there’s another possibility we’re running into here with Kaufman County not fitting with very much. In fact, if Dallas County and Kaufman County are both 54a whole counties, in other words, they don’t get cut, it looks like there’s only one way to go with Kaufman County, and that goes off to VanZant and Rains, but still there’s some policy issues that have to be decided here as to which county can be cut to fix the Ellis County issue, but I just bring this up to highlight especially in east Texas, there are only a limited number of ways you can assemble the counties and the choice you make in one place in east Texas may [a]ffect your options somewhere else, so this is all a big puzzle, especially in east Texas, and a lot of the districts are interconnected and it will [a]ffect things in ways you couldn’t possibly imagine.” D-590 at 47-48. (D) As to whether surplus population could be split between two or more districts, Hanna said, “we don’t have a definitive answer for [that], but I would think that the cautious approach would be to take the entire surplus and put it in with a number of other whole counties. Some people have asked can we split the surplus into two different districts. There are some past plans [in the 1980s] that did that. I think they have some level of vulnerability. I think that goes against the idea of the County Line Rule to not fragment counties, but again, that would be a choice for the legislature to make, but I think it presents a higher risk if you do fragment the surplus into more than one district.” D-590 at 51. Hanna noted that “the language of the constitution does speak to a singular, take the surplus as if it was just one.” Id. at 51-52. But he 55a noted that it was written before one-person, one-vote had been applied, and “the way it’s written in the [Texas] Constitution is not the way it works today because the people that wrote the Constitution didn’t know there was going to be this One Person, One Vote Rule coming, so I would be reluctant to make too much into, you know, reading the singular and they just knew this. They actually were going to draw the districts a whole different way, where they would draw what was called a flotorial district, where they would take the whole county and attach it with another county. That has since been invalidated under One Person, One Vote, so I’m not sure that I would make too much out of the singular in the Consti- tution because the thing they suggest in there may not work anyway.” Id. at 54. He also acknowledged that not splitting the surplus was consistent with the policy of not fragmenting counties. Id. at 56-57. (E) Archer also testified. He stated, “For purposes of minority representation and avoiding a violation of federal law, that is, the Constitution equal protection provision, and more importantly, the Voting Rights Act, you do have to analyze your population in terms of its ethnicity. The Census is insufficient really to do a good job of doing that, so first of all, you have to get out of your mind that if you can draw a district that’s 51 percent black, that is a good black district or a required black district or a non- retrogressive black district or performing, or anything that you’re going to, you know, 56a any terminology or threshold that you’re looking for. Census total population is not going to tell us what we need to know.” D- 590 at 63-64. “[T]he Fifth Circuit Court of Appeals has consistently said you consider citizenship or what is the business now called CVAP, Citizen Voting Age Popula- tion, and so the Census, in itself, does not include citizenship, as you know. There’s some citizenship data that’s available on a five year average that we can do some – we will, in fact, use to help give you some tools to help you.” Id. at 67. Hanna explained that CVAP data would not be available in RedAppl but would be in a separate report, and that they were working on getting an acceptable margin of error for the data. He said, “[W]e decided this was just too compli- cated to put in RedAppl, we didn’t want to hold up RedAppl and we weren’t sure about the accuracy, and still aren’t, to be honest with you, the accuracy of the data to begin with. We’re pretty much the first state out having to deal with this issue, so there’s nobody else to look to, so we’re pretty much having to navigate this by ourselves.” Id. at 71. Hanna also stated that he was unaware of another tool to “get us there.” Id. at 73. An unidentified speaker stated, “Talking about this as an essential tool, while the Fifth Circuit and other Courts have focused on the need to take into account non- citizens, they don’t have any more accurate data either, so what they’ve done is strug- gled to say, well, if you’re barely there and we have some evidence of non-citizens 57a based on your last Census or other things, there are proxies for it, I guess, based on what data is available. When you’re drawing a district and you’re attempting to ensure that minority voters have a reasonable opportunity to elect candidates of their choice, is what Section 2 of the Voting Rights Act guarantees, there are proxies for determining whether they have that based on pure statistics, because it really is, even the statistics are just the starting point. You have to look at political reality, to some extent, cross-over voting, turn-out rates, and there are some gray areas here. It’s not totally clear when massive turn-out by non- minorities in a polarized voting situation means you have to raise that number way above the citizen VAP majority. Does that mean the minority group, they have an opportunity, they’re just not – they’re not using it, or they really don’t have an opportunity? So you can see there are a lot of tricky combination of fact and law issues that are very regional, change over time. Some of the other, as I say, proxies, electoral success in, itself. Even in the Thornburg v. Gingles case, some of the districts were held, multi-member districts were held to be valid because minority candidates were routinely winning and the Courts said we don’t have to go into analysis and counties X, Y and Z because minority candidates are winning, so if you can look at composite election returns for a proposed district and see that minority voters are electing their candidates of 58a choice, in reality, in that district, you have a good proxy for some other more technical measures, such as citizen VAP. Another thing is, as you know, and it’s, again, different people will use it in different ways, but we also have in Texas available Spanish surname voter registration by precinct. Again, we can do allocations. They’re not always perfect because you’ve split precincts in your plans, quite frequently, and the election precincts change over time, so when you go back in time to look at those numbers to say see how the district is changing, those numbers may not be exact for your district, and, of course, the way those lists are compiled, my general understanding is according to the 2000 Census, the Census Bureau correlates surnames with Hispanic – with people who respond there that have Hispanic or Latino background, makes that correlation, and they have to be strong correlations. I mean, there are names that are going to be 67 percent of the people with that name say they’re Hispanic, so I don’t know exactly how they make that initial determination, and then address matching and so on is used to take current voter registration and assign the surname of the voters to the precincts. So there’s a lot of room for error in the assigning of those addresses. Of course, voter registration lists don’t necessarily mean – there’s occasional purges and over- registration, and those sorts of things as well, so it’s not precise, but you can look at those numbers compared to other districts 59a and at least use them as a tool to estimate whether, quote, the minority population is large enough to win elections, to have a reasonable opportunity.” Id. at 75-77. (F) With regard to preclearance, Archer stated, “In establishing that benchmark, the Jus- tice Department is going to start by looking at the voting age populations, minority, non-minority, of existing districts to determine at least a starting point for where is this benchmark that the state has to meet. Their own guidelines, their own anticipated rules, their practices, their letters of preclearance articulate that voting age population under the Census is just a starting point. They’re going to look at overall electoral reality. So again, they’re going to look to see if the districts are, as some people say, performing or effective, as opposed to just a pure mathematical analysis, but they’re going to start with the mathematical analysis, but when people tell you the benchmark is X because I have 23 districts above 46 percent, that’s not reality. And at the same time, you can’t say a 65 percent Hispanic district is, quote, good enough or not good enough in any particular location. It’s a question of political reality. The benchmark is a more subtle, I think, argument in some respects than a Section 2, because the Gingles bright line test creates a hurdle for a plaintiff to establish there could have been more minority districts or there could have been a minority district that you did not draw or the one you drew is ineffective, whereas this benchmark has 60a to do with districts that may be too large in population, may be too small in population. If you have multiple districts that have not grown at the state rate and they are effective minority districts, it may be difficult to maintain them all without making some drastic changes or reducing the minority populations enough so that you can maintain them all versus keeping some stronger, so it’s a little bit more of a juggling act, I think, in that respect.” D-590 at 81- 83. (G) Rep. Villarreal asked, “[C]an we apply the Gingles case to coalitions, so not one minority group but black and Hispanic coalitions?” Archer replied, “I think that the Supreme Court hasn’t given us enough guidance to be absolutely sure. The Fifth Circuit, in some challenges of local govern- ments over the last three decades, has said that an effective coalition of blacks and Hispanics can be the first prong of the Gingles test. What constitutes that coalition on an on-going political basis is a combination of, you know, actual, like I say, history and election, you know, statistics . . . . The Courts are going to look at how those communities work together politically and how they vote, probably most importantly.” D-590 at 85. 97. On March 3, 2011, Interiano emailed Opiela asking, “Do you have block level political data?” US- 61a 104.10 Opiela responded, “yes for 04, 08, 10.” On March 4, Interiano replied, “Need to talk to you first thing tomorrow morning . . . want your help with something using 2008 data.” Opiela responded, “Happy to help” and they arranged to talk later that morning. US-104. 98. On March 7, Interiano emailed Hanna with the subject “Harris County.” D-135; US-157. He asked if Hanna could meet with the Harris County GOP delegation on March 9, but he wanted to talk beforehand “to make sure that we are on the same page.” Regard- ing the number of districts that Harris County would get in the House plan, Hanna wrote, “Arguments for 25: We’ve rounded up before. It fits within the 10% overall deviation. Harris County has a lot [of] votes in the legislature. You have to bring this claim in state court. Arguments for 24: ‘As nearly as may be’ means something, and one number only – not a range of numbers. Very safe from state lawsuit – putting the wrong number in Harris County is a catastrophic error if you guess wrong and requires all of Harris County and most of the rural parts of your map to be redrawn.” Hanna explained that the risks were “fairly small on the substance,” but to redraw from 25 to 24 would require redrawing in Harris County and elsewhere in the state where the other new seat was awarded. TrJ1202. 99. On March 8, Solomons sent a memo to House members stating that he had filed “shell bills” for the various redistricting bills (H.B. 900 for Congress, H.B. 600 for SBOE, and H.B. 150 for the Texas House). D- 145. He encouraged members to be working on their map suggestions and reminded them that the hearing

10 The hearsay objection (TrJ1494-95) is overruled because this exhibit is not admitted for its truth. 62a for the Texas House districts would be Thursday and Friday, March 24-25, 2011 and the hearing on the congressional districts would be Thursday, April 7, 2011. He wrote, “My staff in the Redistricting Commit- tee and I would appreciate your initial suggestions and comments as soon as possible to help prepare for these hearings.” 100. On March 14, 2011, Solomons sent a memo to the members of the House titled “Solicitation of input on Redistricting Bills.” D-145. He reminded them of upcoming hearings and said, “I would like to thank all of those in the Capitol and in the general public who have submitted their comments and suggestions.” The memo asked for cooperation on two matters. First, to be mindful that the staff would be busy after the hearings trying to construct committee substitutes that reflect the realities of the Census, the requirements of the VRA and the Texas Constitution, and the wishes of the elected officials and constituents of the districts. The memo asked that further suggestions or comments be brought to Solomons or to the Committee Clerk, Bonnie Bruce. Second, it informed members that if they had questions about the requirements of the VRA, the Texas Constitution, or apportionment, they had “several resources available” to assist them. These included information on the TLC’s website, which included links to the Texas Secretary of State’s website for historical election data, and the US DOJ’s Voting Rights Section. It said, “For more specific legal questions or to review the legality of possible maps you or constituents in your area may have drawn, please contact the Legal Division of the Texas Legislative Council at 512-[XXX-XXXX] and ask to speak to an attorney on redistricting. The Redistricting Division of the Texas Legislative Council, at [XXX-XXXX], can assist you with questions about RedAppl and can run 63a reports which are not available on RedAppl draft plans, and can print color and larger maps for you at a charge. My Committee Clerk and I will also do our best to assist you and your staff with any questions which cannot be answered by one of these resources.” Id. No mention was made that the Texas OAG or Baker Botts were available as resources. 101. As Committee Clerk, Bruce was in frequent contact with Hanna and Archer at TLC and sought their advice. TrJ1940 (Bruce). Bruce was in charge of setting up a time-line for passing the House map. Bruce wanted the House redistricting bill voted out of committee no later than April 25, preferably April 18, because Easter was coming, and if the bill was not out of committee by then, she felt it was not likely to make it through the entire process. TrJ1939 (Bruce). She worked with TLC and with the House Parliamentarian, Chris Griesel. TrJ1924-25 (Bruce). In doing so, she reviewed what had been done in past redistricting sessions. On March 16, 2011, she emailed Solomons, copying Downton, Interiano, and Griesel, with the subject “Calendar Rule.”11 US-106; Quesada- 249; TrJ1938 (Bruce). Bruce wrote, “1) In 2001, the redistricting bills’ committee substitutes were considered in four hearings, two of which they took amendments (and I mean one bill was heard in five hearings, the committee sub in four and amended in

11 A calendar rule is something the Calendars Committee recommends to the House to govern the debate on the bill. TrJ1203 (Hanna). A calendar rule is proposed by the bill author or committee chair, and then taken to the Calendars Committee. If the Calendars Committee approves it, the Chair of the Commit- tee takes it to the House floor, and the House floor must vote. TrA1082 (Hunter). The House can debate whether a calendar rule should be adopted, and a member can object. TrA1083 (Hunter). There is always a vote on the calendar rule. Id. 64a two of those), so clearly this is a completely different process from how we have done things. A calendar rule was offered and failed, but another calendar rule was adopted; see text below. By the time the bill was considered on the House floor it was on the emergency calendar on May 7th and took all of one day on the floor and was adopted on third reading with little problem the second day. It failed in the Senate. 2) The congressional bills filed in 2003R had 7 public hearings, the committee substitute was heard in six of those hearings. A calendar rule was offered and failed. This bill was heard on Major State on May 12th. In the first called session there were four public hearings, all of which heard the committee substitute. It was placed on the Major State Calendar and no calendar rule was offered. It died on the Senate intent calendar. In the second called session, the bill was filed and the House suspended the rules and voted on the bill on the second day of the special it was reported engrossed and then the Senators fled to New Mexico. The third called session, the bill was considered in a formal meeting of redistricting, there was no calendar rule and it was placed on Major State. The Senate passed the bill.” Bruce sent a follow up email that said, “Excuse me. The calendar rule was adopted in the 2003 R session. The calendar rule was adopted on 5/1/2003. The deadline was to be filed by Noon 5/11/2003 and the bill was eligible at 10am on 5/12/2003.” US-106. 102. On March 17, Bruce sent another email to Solomons, copying Downton, Interiano, Griesel, Dyer, Archer, and Hanna with the subject “RE: Calendar Rule.” US-107; Quesada-246. She wrote, “So I talked to Chris this morning. David Hanna believes the 2001 calendar rule was the better rule (see text below). This covers amendments and amendments to the amend- 65a ments. The 2001 calendar rule was offered on Tues- day, May 1 and allowed for submission of amendments by 2:00 p.m. on a Saturday for a bill being heard at 10:00 a.m. on Monday. This is about the max time we’re gonna have to review these amendments (I have copied Clare [Dyer], Jeff [Archer] and David [Hanna] on this email) because the calendar rule must be offered by the CALENDARS CHAIR, before the calendar which the bill is set on which must lay out for 36 hours and the calendar rule itself must lay out for 6 hours.” She then sent out a proposed time-line and wrote, “As you can see, it’s really about 4 days depending on how quickly we can turn around the committee report documents. The absolute drop dead deadline to get a bill out of committee in order to make the LAST calendar is the bill must be turned into the Committee Coordinator by Thursday, May 5th - that’s 41 days from today. So, we need to decide 1) if we are having public hearings on the committee subs; 2) if you want to have public hearings on the subs at hearings different from what we have already posted . . . ; 3) How far in advance of the public hearings do you want them posted on the internet; 4) When do you want to vote the bills out of committee; and 5) when do you want them on the House floor. Then it’s just a count backwards from the date you want bills on the House floor.” US-107; Quesada-246. 103. On March 18, Bruce sent an email to Denise Davis, Chris Griesel, Solomons, Interiano and others titled “Schedule.” She wrote, “Denise and Chris, Rep- resentative Solomons, Gerardo and I were talking at our last meeting [about] setting a time frame for consideration of the redistricting bills in committee and on the House Floor which would be the least disruptive to the calendars near deadline for passing bills (Chris reminded me last night that we have the budget and 8 66a Sunset bills to get through as well) and which will provide adequate time for the committee to complete their work. Representative Solomons was wondering if this might be done Monday but with Voter ID on the floor, I’m not sure this is possible. Would you be available for a 6pm meeting on Tuesday in Representative Solomons’ office to discuss the best timeframe to move the bills?” D-227. 104. On March 21, Bruce sent an email to Denise Davis, Lisa Kaufman, Solomons, Griesel, Interiano, and Downton titled “Schedule & Issues” with a proposed schedule. D-228; US-108A. The schedule included “Start showing House Map to key members” the week of March 26, and an April 5 announcement by Solomons on the floor that the committee substitute would be released that day. The list of “issues to discuss” included “Harvey Hilderbran, Todd Hunter, Otto/Hamilton, Harris County (although their meeting on Sunday went well supposedly), Branch/Hartnett, [and] Schwertner/Gonzales,” and the list under “Liti- gation team advise needed” included “Nueces County, Vo/Hochberg pairing, and Harris County (24 v. 25).” D-228. This indicates that final decisions on these issues had not yet been made. 105. On March 24, the HRC held a public hearing to solicit input from the public regarding the apportionment of the members of the Texas House of Repre- sentatives. D-116 at 136; D-6; D-592 (transcript). There was no proposed map at this hearing, only a map of the current districts. D-592 at 6. In response to questioning, Solomons stated that he hoped to start trying to put together a map and talking to members about redistricting “starting next week.” Id. at 7. A number of witnesses testified, including Rogene Cal- vert (Texas Asian American Redistricting Initiative, 67a “TAARI”), Luis Figueroa (MALDEF), and Gavino Fernandez (LULAC). D-116 at 139. Jose Garza, attorney for the Mexican American Legislative Caucus (“MALC”), testified that Latinos in West Texas, Dal- las, and the Valley should get additional opportunity districts, and that a majority-minority district could be drawn in the Grand Prairie area of Dallas. D-6. Rep. Veasey felt that this hearing did not allow for meaningful pubic input because no public plan was available. TrJ16 (Veasey). March 25 had been designated for the HRC to hear additional testimony on the House plan if necessary and to consider the SBOE bill, H.B. 600. D-116 at 141, 142; D-593. The HRC held a public hearing on March 25 on the SBOE plan. D-116 at 143. There was no testimony on the Texas House plan. 106. As the mapdrawers were working on plans, Hanna did not provide mapdrawers with any election analysis; instead, he directed them to OAG. TrJ1613 (Interiano). Mapdrawers received election analyses from the OAG, and these analyses were important in their mapdrawing. TrJ1615, TrA5 (Interiano); TrJ1015- 16 (Solomons). The reports from the OAG included comparisons of proposed plans to benchmark using ten statewide general elections involving minority and Anglo candidates between 2002 and 2010 referred to as the “OAG 10.” TrA6 (Interiano); Tr959 (Downton). The elections included were (1) 2002 general election for Governor between Perry (Anglo, Republican) and Sanchez (Hispanic, Democrat), (2) 2004 general election for Railroad Commissioner between Carillo (His- panic, Republican) and Scarborough (Anglo, Demo- crat), (3) 2004 general election for Court of Criminal Appeals between Keasler (Anglo, Republican) and Molina (Hispanic, Democrat), (4) 2006 general election for Lieutenant Governor between Alvarado (Hispanic, Democrat) and Dewhurst (Anglo, Republican), (5) 2006 68a general election for Court of Criminal Appeals between Keller (Anglo, Republican) and Molina (Hispanic, Democrat), (6) 2008 general election for US Senate between Cornyn (Anglo, Republican) and Noriega (Hispanic, Democrat), (7) 2008 general election for Justice of the Supreme Court between Johnson (Anglo, Republican) and Yanez (Hispanic, Democrat), (8) 2010 general election for Lieutenant Governor between Chavez-Thompson (Hispanic, Democrat) and Dewhurst (Anglo, Republican), (9) 2010 general election for Land Commissioner between Patterson (Anglo, Republican) and Uribe (Hispanic, Democrat), and (10) 2010 general election for Justice of the Supreme Court between Bailey (Anglo, Democrat) and Guzman (Hispanic, Republican). D-418. 107. The OAG reports identified the minority- preferred candidate in each district in the plan under analysis and whether the candidate would have carried the district. TrA6 (Interiano). The OAG also created and provided summaries because the reports were very long. TrA7 (Interiano). Interiano testified that these analyses were important to him in analyzing draft plans for VRA compliance. TrA5 (Interiano). Interiano shared reports or information amongst mapdrawers and with people outside the process, like Opiela or others who requested it. TrA8 (Interiano). Bruce received RPVA reports from the OAG and shared them with Downton, Interiano, and Solomons, but not with minority members of the HRC or of the House. TrJ1957 (Bruce). Solomons did not share the analyses with minority legislators on the redistricting committee. TrJ1023 (Solomons). Solomons did not tell any minority legislators on the redistricting committee that the OAG was conducting and providing analyses. TrJ1024 (Solomons); TrJ1233 (Thompson). Solomons testified that if a member had asked to see 69a a copy of the RPVA, he would have let them see it. TrJ1077. When asked how a minority member would have known to ask for the RPVA, Solomons said, “I would have thought that minority members . . would have asked either me or my staff or [Interiano], ‘Do you have anything I can work with?’ or whatever, and they would have said, ‘Yeah.’” TrJ1088. 108. On March 25, Stacy Napier of the OAG emailed Downton with the subject “reports.” US-158. She attached various reports and wrote, “[H]ere are the reports we discussed. As I mentioned, LTS (our Legal Technical Support Division) chose 10 contested statewide general elections spanning 2002-2010 and determined who the Hispanic candidate of choice was in each of those elections through their regression analysis. Under the current map, Hispanics in District #1 were able to elect the candidate of their choice 7 out of the 10 times. Under the proposed map, Hispanics in District #1 would be able to elect the candidate of their choice 3 out of 10. The summary chart is the excel file attachment. I have also asked our LTS division to run reports on a Midland/Ector switch to see how that looks.” US-158 (emphasis in original). 109. On March 28, Interiano emailed Speaker Straus, copying Denise Davis and Kaufman with the subject “Recommendation.” US-159. He said “it might be best for you and Mr. Solomons to visit on the changes that I made to the map since our first conversation and get his input before you visit with other members. He may also be able to help you visit with those members once you are both comfortable with the map. I also wanted to let you know that I never showed Mr. Solomons the map that I presented to you last week. That map was just based on some of the guidance that I had been given but you were the 70a first one to see it. With that said, all of these maps were done with information and input from his staff on our meetings with members.” 110. Downton’s RedAppl account begins to include congressional plans in late March. D-508. April 2011 111. On April 3, Lamar Smith emailed Denise Davis stating, “We have all benefitted from talking w tom phillips and sam cooper. Wld u thank the speaker for making them available. And we will appreciate their taking a look at our proposed map in the next day or so. Also I think it will be of interest to all of us how the clearance and appeals process is likely to unfold. For ex we agree that we are not going to seek doj preclearance but will go to a three judge panel in dc. Who is likely to be picked. (Three will be chosen from close to 30 two thirds of whom have been appointed by dem presidents.) This is another reason why map needs to be balanced if we are going to avoid having a court draw the map. The panels decision will be appealed by some party. Who is likely to hear appeal. Then will it go to scotus. How long will it take etc. Could tom and sam give you a memo on this process with some strategic advice that might be helpful in persuading st reps and others to support our proposal.” NAACP-61. Davis responded to Smith, “I will check on this tomorrow and get back to you.” Davis forwarded it with the comment, “Thoughts?” and Kaufman responded, “Need to discuss,” copying Interiano. US-115. Interiano responded, “Agree with Lisa but think that we should be really careful in using anything like that to ‘persuade’ members. This email is Lamar’s way of letting us know that his map is a 2- 2 map. I’ll be curious to see if Eric even gives us a 3-1 as I had suggested that he do to give us options.” Id. 71a 112. On April 5, 2011, Lamar Smith’s office sent Interiano a confidential congressional redistricting proposal. It states that it: “Maintains the core of current districts unless requested otherwise; Strives to be fair and reflect the changing demographics of Texas; Creates four new districts as allowed by the census results 1. One new likely Republican district in East Texas, 2. One new likely Republican district in South Texas, 3. One new Voting Rights Act district in South Texas that leans Republican, 4. One new Voting Rights Act district in the Dallas-Ft. Worth area; Results in 25 congressional districts that lean Republican and 11 that lean Democratic.” PL-311; NAACP-613. 113. On April 6, Opiela emailed Downton a proposed congressional map at 2:36 p.m. PL-311; D-608. The email subject is “0406” and the attachment is “0406.zip.” At 3:54 p.m., Downton sent back a pdf file called “Congressional 1.pdf” and the attachment is a statewide map of hrc1C104. The hrc1 RedAppl plan log shows hrc1C104 as an imported map created April 6 at 3:51 p.m. D-508.1. 114. The HRC held a hearing at 9:00 a.m. on April 7, 2011 to hear invited and public testimony on potential Texas congressional districts in light of the 2010 Census data. D-116 at 111; PL-465; D594 (transcript). The hearing took place in the Capitol Extension Audi- torium, while the House was in session. D-116 at 112; D-594. There was no draft congressional map available for the hearing. D-594. The hearing lasted about four hours, and witnesses testified. D-8. (A) Rep. Alonzo (Hispanic, Democrat) advocated for a minority district in the DFW area. Rep. Menendez (Hispanic, Democrat) read a statement on behalf of Congressman Charlie Gonzales that advocated for the 72a creation of a new minority opportunity district around the San Antonio area while not disturbing the traditional lines of the existing minority district CD20. Rep. Dawnna Dukes (African American, Democrat) advocated on behalf of Congressman Lloyd Doggett for the reunification of Travis County and Doggett’s congressional district. She noted that Austin and Travis County voters had formed a coalition to elect minority candidates of choice, and that African-American voters strongly supported Lloyd Doggett and would like to be reuni- fied in his district. Former Austin Mayor Gus Garcia also advocated for having all of Austin in one congressional district. A number of other witnesses also testified in favor of not dividing Austin among several districts. Lloyd Neal, Nueces County Judge and former mayor of Corpus Christi, advocated for a new congressional district being formed around Nueces County/Corpus Christi and moving to the north and perhaps the west. D-594 at 17. Matt Shahan of the Collin County Commissioners Court and Keith Self, Collin County Judge, advocated for a congressional district wholly within Collin County and not connected to Dallas. Id. at 19-22. Tarrant County Com- missioner Roy Brooks testified that the African-American community of southeast Fort Worth had been harmed in the 2003 redistricting by being placed as an appendage to Denton County in CD26, and asked that the community be reinstated as the core of a district made up of Tarrant and 73a Dallas County voters. Id. at 31-36. Judy Cope of Guadalupe County presented a resolution of the county commissioners court asking to be taken out of CD28 and placed into a district consistent with Guadalupe County’s historical voting patterns and with other counties of similar size. Id. at 54-55. Sergio DeLeon and others advocated for a Latino opportunity district in North Texas/DFW, and Rep. Peña (Hispanic, Republican) expressed his support for such a district. Id. at 60-6. A Senator from Houston advocated for a new Latino district there given the Latino growth. Orlando Rios testified on behalf of Congressman Cuellar and asked to keep Webb County in CD28. Id. at 69. Congressman Flores’s district director asked that CD17 be kept as close to its current configuration as possible. Id. at 70-71. (B) Nina Perales testified that two new minority opportunity districts were required and proposed maps C108 and C109 from the Texas Latino Redistricting Task Force. The maps contained nine Latino districts, including a new district in DFW. Plan C108 created a district that combined the Latino populations of south San Antonio and southeast Austin. Id. at 78-80. 115. On April 7, Hanna wrote the first of three memos on draft House plans; the first was entitled “Possible Retrogression Issues for Black and Hispanic Districts in Proposed House Plan.” D-122; US-347. Hanna was asked by Bruce and Downton to do a retrogression analysis on the House plan in progress 74a (drh1H109 in his plan log). TrJ1157, TrJ1219 (Hanna); TrJ1941(Bruce); TrJ2000 (Downton); D-325 (Hanna plan log). Relying on total SSVR for Hispanic districts and Black total population and BVAP (B/BVAP) for African-American districts (TrJ1187), Hanna made the following points: (1) the decrease in B/BVAP in HD22 (Rep. Deshotel) “below 50% indicates further election analysis should be conducted to make sure the district still performs and whether a better-performing district could be drawn.” (2) “Nueces County may be the single most difficult retrogression issue to predict. While there are two 50% SSVR plus districts within the county currently that may constitute performing Hispanic districts, they are both significantly underpopulated and the remaining people in Nueces County are predominantly Anglo. The county line rule likely requires two districts to be wholly contained within Nueces County with no surplus coming out; however this would have to yield to the federal Voting Rights Act if it can be shown retrogression could be avoided by splitting the county. The approach taken in the proposed plan is to draw one clearly performing district and one that is not. Another approach is to split the Hispanic population exactly in half, resulting in two districts that are slightly at or under 50% SSVR, though neither will likely reliably perform as Hispanic districts of choice. A final approach is to see if by splitting county lines in the area, the second Hispanic district could be preserved. This approach should be further investigated though it runs 75a the risk of violating state law and requiring other county lines to be split, and should be pursued only if it would clearly contribute to total Hispanic voting strength statewide.” (3) “HD35 drew an objection from DOJ for having its SSVR lowered from 55.6% to 50.2% and pairing an Anglo and Hispanic rep. in a district DOJ concluded favored the Anglo. The court remedy only increased the SSVR to 51.5 but drew the district as an open district that a Hispanic candidate could win. The proposed plan lowers the SSVR 2.5 points but has it higher than the court plan 10 years ago. Further election analysis should be performed on the district to measure the performance of the existing and proposed districts.” (4) “In 2001 a Cameron County district drew an objection from DOJ for lowering the SSVR in Dist 38 from 70% to 60%. In the proposed plan the Cameron County districts appear to not be retrogressive, while the Hidalgo County districts are more problematic. In particular District 40 takes a significant dive in SSVR from its current level. The significant rework- ing of the district lines in Hidalgo County means that new District 40 is likely more comparable to old District 41, but even here the SSVR drops by almost 6 points. With all adjoining districts having substantially higher SSVR there is a significant risk that DOJ will conclude that the change in Dist. 40 is retrogressive with regard to Hispanic voters through packing of Hispanics in the other districts. The safer approach would be to restore Dist. 40’s SSVR to the previous low 76a SSVR for a district in Hidalgo County which was 68.7 %.” (5) “Even though Travis County probably has one of the lower instances of racially polarized voting in the state, the drop in the SSVR in District 51 seems to present a need- less risk for retrogression that might be avoided by some simple precinct swapping.” (6) “While most of El Paso County does not appear to have retrogression issues, the slight decrease in SSVR in District 78 [from current 47.1 to proposed 45.8] does present some level of risk of retrogression that likely could easily be remedied by swapping some precincts with an adjoining district. Section 5 analysis frequently focuses on the differences in minority voting strength in adjacent districts.” (7) “While current District 90 [Tarrant County] is short people and that likely accounts for most of the drop in SSVR, further consideration should be given to see whether the level of SSVR in the proposed plan can be raised to come closer to the level in the current plan.” (8) “These two Hispanic districts in Dallas County [HD103 and HD104] both present retrogression issues. While both are significantly short people (more than 86, 000 combined), no new Hispanic districts are being added in Dallas County, and the overall percentage increase in the Hispanic population in Dallas County makes the declines in SSVR especially difficult to justify. While it can be argued that District 104 will likely perform at 77a 45.6 SSVR since this is similar to the performing level it was drawn at in 2001, no similar argument exists for the reduction in District 103. Accordingly, at a minimum, the decline in SSVR in District 103 should be remedied. Consideration should also be given to keeping District 104 over the 50 % SSVR threshold if this can be done.” (9) “The Black population in Dallas County is moving to the south and West and out of the inner city. This is reflected in the disparities in the Black populations in the current districts between the two more northerly ones (100 and 110) and the two southern ones (109 and 111). This effect is enhanced by the proposed plan and presents clear retrogression issues. With the Black percentage in District 109 sitting at 62.1% in the proposed plan, it is likely that the levels of Black populations in District 100 and 110 can be restored significantly closer to their current levels without endangering the viability of District 109 or District 111 as performing Black districts. Additional leveling out of the Black populations could occur, but likely would not be required for preclearance under Section 5.” (10) “Bexar County’s seven Hispanic districts may constitute one of the most challenging balances of population in order to avoid retrogression. In 2001, the state proposed eliminating one of these districts because of the loss of a district in the county but preclearance was denied for this proposal by DOJ. In its fix, the court chose to draw seven Hispanic districts each with an SSVR of 55%. In the 78a decade since the court drew its plan, six of the districts have remained at the level of SSVR at which they were drawn, and one (District 117) has diminished to just above 50% SSVR. Bexar County as a whole has remained relatively constant in SSVR over the decade going from 42.4% to 43.1%. Despite this near constant level of SSVR, it is possible that His- panic voters have become more dispersed across the county making Hispanic districts more difficult to draw. Five of the seven districts are short people necessitating a move of districts to the north and west. The proposed plan raises retrogression questions as to the significant declines in Districts 116 and 119. These declines seem inexplicable in light of the raises in the SSVR levels in adjacent Dis- tricts 118 and 125. The declines in SSVR in Districts 123 and 124 seem to be only of minor concern. The most prudent approach would be to eliminate the increases in SSVR in Dis- tricts 118 and 125, and restore as much of the declines as possible starting with Districts 116 and 119.” (11) “While most of the performing Black districts in Harris County do not appear to have retrogression issues, the decline in District 147 in the proposed plan may be a problem. While three of the other Black districts (139, 141, 142) see increases in their Black populations over their current levels, the Black district with the lowest percentage of Black population (147) sees a decrease. This would appear to be a potential retrogression issue. The solution would be to increase the Black 79a percentage in District 147 so that it does not decline from current levels.” (12) “Of the four performing Hispanic district[s] in Harris County, three appear to be in pretty good shape in the proposed plan. Only District 148 is dropping in its level of SSVR. Unfortunately this district is the lowest in SSVR already so there is a possible retrogression issue with District 148. Given that there is a sizable cushion in adjacent District 145 over 50% SSVR, it should be possible to restore District 148 to its current level of SSVR.” (13) “When examining the overall current Hispanic population percentage (63.8) or even the HVAP (59.8) one might assume that District 137 is an effective Hispanic district. However because of the high number of non- citizens in the district, the SSVR is in the low 20’s and as such this is not currently a performing Hispanic district. In any event the SSVR increases so there is little possibility of retrogression with this district.” (14) “This multiethnic district in Harris County [HD149] is eliminated. The ‘other’ population is primarily Asian but represents different ethnic groups. If it can be determined that the district was a true minority coalition district, there could be a retrogression issue in its elimination but this would be a novel retrogression theory to apply where no single racial or ethnic group has more than a quarter of the VAP of the district.” 80a Hanna did not think HD149 in Harris County was a protected district. TrJ1158 (Hanna). Hanna’s memo included only demographic analyses and no election analysis. TrJ1188 (Hanna). Bruce shared Hanna’s memos with Interiano, Downton, and Solomons, but not with minority legislators because “they were advice on working documents.” TrJ1957 (Bruce). 116. On April 7, Rep. Garnet Coleman (African American, Democrat) sent a memo to Rep. Beverly Woolley (Anglo, Republican) and the other members of the Harris County delegation about the Harris County map. US-266; TrJ1301-02. Led by Woolley, the 13 Republican members of the Harris County delegation had approved a 24-district map pairing Reps. Hochberg and Vo and provided it to Downton. When Coleman learned about this map, he sent the memo complaining about the draft map with 24 seats and asserted that they should be working on a 25-district map. 117. On April 8, Bruce emailed Hanna, Dyer, Solomons, Interiano, and Denise Davis with the subject “schedule.” D-226. She wrote, “Ryan and I have been talking about schedule. If we can get West Texas, Dallas, Houston, San Antonio and Hidalgo Co. situ- ated by Noon on Monday, we could get Clare to turn around a state wide map with the large county blow- ups, and a 100, 200, and 350 report for [Solomons] to show members of the committee individually on Monday afternoon. All day Monday we continue to show the remaining members their districts (we will all have to wear bullet proof vests that day). Monday night, we put out a press release to be embargoed until Tuesday morning that the maps would be released Tuesday. We make the maps public Tuesday and suspend to meet on Thursday (I assume the SBOE bill will be upon the floor Wednesday) and then we can 81a have a formal meeting during session the following Monday? What do you think?” 118. Later on April 8, Hanna emailed Denise Davis with the subject “slowing down plan a bit.” US-117; TrJ1155. Hanna wrote, “[Slowing down plan a bit] is a good idea. your minority districts have some issues with Sec. 5.” Davis responded, “We talked about that today.” Hanna replied, “Most of that was fairly easy stuff. Later I’ll need to talk to you about the harder stuff, which can’t be fixed in the political process. You and SJS [Speaker Joe Straus] will need to know. I still need to do some work and have JSA [Jeff Archer] review.” 119. On April 12, Hanna wrote a second memo entitled “Possible Retrogression Issues for Black and Hispanic Districts in Proposed Plan H110.” US- 339;TrJ1160 (Hanna); D-327; PL-1620; Quesada-258. This memo again raised retrogression issues with respect to HD22, Nueces County, HD35, Hidalgo County, HD51, HD90, HD103/HD104 in Dallas County, Bexar County, HD148 in Harris County, and HD149, but no longer identified an issue with HD78 in El Paso because the SSVR has been raised to 46.8%. D-327 at 3 (“The minor decreases in the SSVR in Districts 77 and 78 will not likely present retrogression issues.”). With regard to the minority districts in Harris County, Hanna added the sentence, “Consideration should also be given as to whether a fifth majority Hispanic district could be drawn in Harris County and whether such a district would be required by Section 2 of the Voting Rights Act.” Id. at 5. Hanna had been able to draw a fifth district in Harris County, but was not sure it was “politically viable.” TrJ1207 (Hanna). 120. The five-day posting rule was suspended on April 13, 2011. D-116 at 101. Villarreal (Hispanic, 82a Democrat), vice-chair of the HRC, made the motion to suspend the rules. TrJ1944 (Bruce); TrJ29 (Veasey); TrJ1681 (Hochberg). This allowed hearings to be held on the map on Friday, April 15 and Sunday, April 17. TrJ29 (Veasey). Waiving the five-day posting rule made it difficult for members to sufficiently analyze the proposed plan and generate public response. TrJ1656 (Hochberg). However, it was not necessarily unusual to suspend the five-day rule. TrJ1681 (Hochberg). The Legislature suspended the five-day posting rule in 2001 for redistricting, and may have done so in 1991. TrJ1683-84 (Hochberg). But they did not waive it for every hearing. TrJ1685 (Hochberg). 121. On Wednesday, April 13, it was announced from the House floor that there would be a public hearing on the House plan on Friday, April 15. D-116 at 97. Rep. Sylvester Turner (African American, Dem- ocrat) went to the back microphone and stated that it was very short notice to inform people and have them come to testify or provide public comment. Tr796 (Turner). Turner testified that many members voiced objections that the time-line was too quick and that 48 hours for the first hearing after the map was released was an expedited time-line that was much shorter than in the two previous redistricting cycles, which had given at least a week-and-a-half notice. Id. 122. On April 13 at 11:23 a.m., Hanna emailed Bruce stating, “I think the hearing schedule is a little too tight to receive meaningful input from people. The House will be in session Friday and this will interfere with members being able to come. I would wait longer and not have the hearing when the house is in session. You can still have it out of committee before the Easter break if that is what you want.” D-260. After Bruce responded with concerns about getting the bill out, 83a Hanna suggested having the hearing on Sunday afternoon to resume Monday morning if necessary, with amendments filed by 5 p.m. Tuesday and committee vote Wednesday night or Thursday. Id. Hanna noted that “[t]he process on this is important for both preclearance and litigation. People and members must be given a meaningful opportunity to comment on the plan you lay out. Since I don’t think the senate will do much with it, the only opportunity for pubic input on the plan to be reflected with amendments is in the house committee phase.” Id. Bruce forwarded Hanna’s comments to Interiano, Solomons, Downton, Denise Davis, and Kaufman. Id. As a result, the HRC posted hearing notices for Friday and Sunday, with an opportunity for testimony to roll over to Monday, if necessary. TrJ1944 (Bruce). It was unusual to provide an opportunity for public testimony on a weekend. TrJ1944-46 (Bruce); TrJ1652 (Hochberg). 123. On April 13, Solomons issued a press release stating that the proposed House map (C.S.H.B. 150, Plan H113) would be released at 4:30 p.m. and that there would be hearings on Friday, April 15 and Sunday, April 17 (which was Palm Sunday). PL-205; D-9; US-457. The press release lists the following “key points” for the plan: (1) under “compliance with one- person, one-vote,” overall deviation was within +/- 5% with the average deviation being 2.68% or 4498 people; (2) under “compliance with the Voting Rights Act,” “In Nueces County, which has a 49% Spanish Surname Voter Registration and an apportionment of two districts, one new performing majority Hispanic district is created with a 63.7% Spanish Surname Voter Reg- istration. Maintained all current Black opportunity districts.”; (3) under “compliance with the Texas Constitution’s County Line Rule,” “There is only one county line split, in Henderson County, which was 84a necessitated by the population size of Ellis County.”; (4) regarding “pairings,” the release noted that (a) lack of growth in East Texas led to the pairings of Republi- cans Flynn/Cain and Ritter/Hamilton, (b) lack of growth in portions of West Texas led to the pairings of Republicans Landtroop/Perry and Chisum/Hardcastle, (c) lack of growth in Dallas County led to the loss of two seats (16 to 14), all of the Democrat districts are protected by the Voting Rights Act, which led to pairings of Republicans Driver/Burkett and Harper- Brown/R. Anderson, (d) due to lack of growth in Harris County, the County went from 25 to 24 seats, which led to the pairing of Democrats Hochberg/Vo, and (e) due to lack of growth in Nueces County, the County went from having 2+ districts to 2 districts, which led to the pairing of Republicans Torres/Scott; (5) regarding “open seats,” the map creates eight open seats, HD3 (Montgomery and Waller), HD12 (McLennan, Brazos, Falls, Limestone, Robertson), HD33 (Collin and Rockwell), HD85 (Fort Bend, Wharton, and Jackson), HD88 (Wise, Cooke, Jack, Young, Throckmorton, Haskell, Stonewall, Kent, Garza, Lynn, Terry, Borden), HD101 (Tarrant), HD106 (Denton), and HD149 (Williamson, Burnet, Milam). PL-205; D-9. 124. The first public plan for the Texas House, Plan H113 (Solomons Statewide Proposal a/k/a Committee Substitute to H.B. 150 (C.S.H.B. 150)), was released on Wednesday, April 13 in the afternoon. TrJ1939 (Bruce); PL-205; D-145. Before the map was released, no one saw the statewide version, but individual members were shown their districts or delegation maps. TrJ1942 (Bruce); D-229. Solomons’ office was still working on the map even after it was released. They were still working on West Texas, and neither Dallas nor Harris County had provided an agreed map (Plan H113 used the 24-district map provided by the Harris 85a County Republican delegation that paired Hochberg and Vo and eliminated Vo’s district HD149; the Dallas County map was drawn by Downton, not the Dallas County delegation). TrJ1940 (Bruce). Plan H113 significantly changed Hidalgo County in an effort to create a favorable district for Rep. Aaron Peña. Plan H113 had only 28 districts that were majority SSVR (using non-suspense SSVR), whereas the benchmark had 29. This is because Plan H113 eliminated HD33, an SSVR-majority district in Nueces County, and did not create any new SSVR-majority districts. 125. On Thursday, April 14, Hanna sent an email titled “performing districts” to Interiano stating, “Talked to Archer about the AG thing re ‘performing.’ This is probably good news but the standard is an opportunity to elect candidates of choice so I don’t know exactly how the AG definition of ‘performing’ relates to this. What is their definition of opportunity to elect? Is a performing district a smaller subset composed of districts that consistently elect candidates of choice? Even if we keep the number of ‘performing’ the same, we can still get dinged for eliminating some of the inconsistent ones.” D-138; PL-1618; US-123; Quesada-255. 126. The HRC held the first public hearing on the House plan (Plan H113) at noon on Friday, April 15. PL-206; D-10; PL-224; D-365; D-595. Reps. Solomons, Villarreal, Alonzo, Alvarado, Aycock, Branch, Eissler, Geren, Harless, Hilderbran, Hunter, Keffer, Madden, Peña, Phillips, Pickett, and Veasey were present, although Solomons left during the hearing. D-116 at 97. (A) Villarreal, acting as chair, offered a complete committee substitute and recognized 86a Solomons to explain the committee substitute (Plan H113). Solomons explained that he tried to limit pairings as much as possible and only paired Republicans with Republicans and Democrats with Demo- crats because it was his goal to give each member the opportunity to win the district. D-595 at 5. He also noted that the eight new districts that resulted from the pairings were primarily anchored in suburban counties throughout the state (Montgomery, Brazos, Collin, Fort Bend, Wise, Tarrant, Denton, and Williamson). Id. Solomons noted that the map was a starting point and a “work in progress.” Id. at 5, 8. (B) Luis Figueroa of MALDEF testified in opposition to the map, offering alternative Plan H115. D-595 at 10. Figueroa stated that Plan H113 raised concerns under the VRA because the benchmark had 29 districts with a majority SSVR, and H113 only had 28. In contrast, he asserted, Plan H115 created 34. He stated that H115 had 35 HCVAP-majority districts (the benchmark had 30) and created five new “citizen voting age population districts”—one in the Pan- handle, one in West Texas (Midland), one in Hidalgo County, one in Tarrant County, and one in Houston. Id. at 61-62. He also asserted that the loss of the Latino opportunity district in Nueces County raised retrogression and § 2 concerns, and that the failure to create a new Latino opportunity district in the Rio Grande Valley raised § 2 concerns. He further contended that the map packed Latino voters in El Paso and 87a Harris County to create fewer Latino voter registration majority districts. In response to questioning about the County Line Rule, Figueroa admitted that the map broke the rule in the panhandle and West Texas/Mid- land to create § 2 districts, but asserted that this was necessary to comply with the VRA. Id. at 12-15, 63-64. Figueroa noted that Cameron and Hidalgo Counties both had excess population that could be joined to create a new Latino majority district. Id. at 24. In response to questioning, Figueroa also pointed out that in HD90 in Tarrant County, Solomons’ proposal reduced the (non-suspense) SSVR from 47.2% to 41.9% while MALDEF was able to increase it by almost ten points over Solomons’ map. Id. at 32. He said MALDEF wanted to “ensure that that district becomes a Latino citizen voting age population” majority. Id. at 66. Further, he pointed out that in HD148, “[SSVR] is also reduced whereas we were able to even out the population to increase Representative Farrar to make it a Latino majority district.” Id. at 32. He also noted that El Paso did not have an even distribution of Latinos and “we could definitely add Latinos to District 78.” Id. Figueroa stated that the map did not reflect minority population growth and violated both § 2 and § 5. Id. at 33. Figueroa asserted, “we can draw Section 2 compliant districts in West Texas, in the Valley, and increase populations in Houston and Representative Farrar’s district [HD148] and in Lon Burnam’s district [HD90] to create a clear Latino CVAP 88a majority district.” Id. at 34. Plan H115 also attempted to maintain HD35 as a Latino district. Id. at 64-65. Figueroa testified that the County Line Rule should yield to create minority districts. Id. at 94-95. He said that the other county cuts were made to comply with one person, one vote, and they would be happy to re-evaluate those. Id. at 96. Figueroa said that districts 81 and 87 were compact given the lower populations in the area and did not split any precincts. Id. at 108-116. Rep. Peña asked “Other than your assumption that District 32 ought to come into Hidalgo County and, you know, going around the county line rule, is that the only objection you have to Hidalgo County as violating the Voting Rights Act?” Id. at 123. Figueroa responded, “Yes, I think that’s right. . . . I think that we only wanted to create one more Section 2 compliant district.” Id. at 123-24. Figueroa agreed that the rest of the map in Hidalgo County was “compliant with the Voting Rights Act.” Id. at 124-25. (C) Rep. V. Gonzales (HD41), Rep. Martinez (HD39), and Rep. Muñoz (HD36) of Hidalgo County testified against Solomons’ plan. Rep. Gonzales submitted a proposed amendment that would keep the core of the current districts while increasing Republican precincts in Peña’s district and was approved by Muñoz, Martinez, and Gonzales. D-595 at 135. (D) Rogene Calvert of TAARI testified against Plan H113 and the pairing of HD137 and 89a HD149 in Harris County. D-595 at 181. She said the Alief ISD was split among five districts and the Asian vote was diluted. Id. at 182. Other witnesses also testified against the loss of HD149 and to cohesion among minority voters in the area. Calvert also complained about HD26 in Fort Bend County, which had one of the largest Asian- American populations and they had hoped for two more compact districts to protect their strength. Id. at 184. Sandra Crenshaw from Dallas testified against Plan H113, asserting that it improperly split cities. D-595 at 206. She also noted that splitting precincts causes costly problems for elections and confuses voters. Id. at 208. (E) Peña asked when they would be voting on amendments, and Villarreal said, “Right now we’re about to close on the house bill. The chairman is not here so we’re not going to consider any amendments today. And my understanding is that Representative Solomons is not going to be with us on Sunday. And so I think on Sunday it would be a fine time to have members present any solutions, any amendment they want to, but we will not adopt any on Sunday. [Solomons] doesn’t plan on voting Sunday. I understand he plans on voting a House Bill 150 out of committee on Tuesday.” D- 595 at 218-19. An unidentified speaker stated that they were “trying to resolve some issues in Harris County” and requested another day or two beyond Tuesday. 90a Villarreal said to take it up with Solomons. He then withdrew the committee substitute and left H.B. 150 as pending business. Id. at 219-20. 127. On April 15, Solomons issued a press release regarding Plan H115 stating, “I have come to the conclusion that it violates the Texas Constitution. In their singular goal of increasing the number of Hispanic seats they have unnecessarily crossed 32 number of [sic] county lines. The Texas Constitution requires if a district can be drawn within a county it must be to ensure our counties have dedicated representation and their voting strength and representation is not diluted. This only reinforces the traditional legal redistricting principle of communities of interest.” MALC-50. 128. On April 15, Interiano forwarded an email titled “Rep Sylvester Turner - precincts” to Mike Hull (lawyer for the Houston Republican delegation), writing, “Try to see if you can make as many of these changes [to Houston] as possible but again without touching.” Quesada-68 at 21. Hull responded, “Scott [Sims] is working on these.” Interiano replied, “Per- fect. After he’s done with that see if he thinks he can get district 90 [in Tarrant County] above 50% SSVR under the same parameters.” Id. 129. On April 17, the Doggett Campaign sent an email to Rep. Eddie Rodriguez about a press conference on Travis County redistricting to take place on Monday April 18. D-681. It included information about a front-page article from April 16’s Austin American Statesman called “Will Doggett be the target of a new congressional map?” Id. The article stated, “[Lamar] Smith, who is close to Texas House Speaker Joe 91a Straus, has taken the lead among the state’s Republi- can congressional delegation in drawing a map for state lawmakers to consider. The Washington bureau of The Dallas Morning News reported last week that the delegation just submitted that map to state legislators.” Id. 130. On Sunday, April 17 (which was Palm Sun- day), the HRC held the second public hearing on the House map. D-116 at 101; D-11. Chairman Solomons was not present. Villarreal, as acting chair, laid out Plan H113. Testimony was taken. (A) Hanna noted that the DOJ issued new § 5 rules on April 15. Rep. Hochberg testified in opposition to Plan H113. He disagreed with the decision to round Harris County down to 24 districts, noted that the map split the Sharpstown community of interest, and submitted letters in opposition. (B) Rep. Mallory Caraway (African-American, Democrat) representing HD110 in Dallas County opposed Plan H113. Rep. Eric Johnson (African-American, Democrat), who represented HD100 in Dallas County, testified in opposition to Rep. Caraway’s proposed changes that would swap areas between HD100 and HD110. Sandra Crenshaw (Precinct 3549 Chair) from Dallas testified against Plan H113. She stated that Oak Cliff was split among districts and the African-American community in HD110 should be kept together. She objected to suburban districts coming in and “cherry picking” population. She stated that Dallas County representatives had testified about lack of transparency and 92a encouraged the mapdrawers to respect communities of interest. (C) Kevin Burns, Wise County Commissioner, testified in opposition to Plan H113. He complained that the proposed HD88 was too wide, encompassing 12 counties, and included Wise County with areas with which it had no community interests in far west Texas. He asked that Wise County be included more with the DFW metroplex. (This was later changed, and Wise County was put with Parker County in HD61 in the enacted plan). (D) Jacquie Chaumette, at-large council member for Sugar Land (in Fort Bend County), testified in opposition to the map. She complained that precinct 4102 was split and the portion taken out of HD26 (the Avalon subdivision) had been paired with Wharton County in the new district HD85, which is not a community of interest. She stated they had just gotten notice on Friday and the community was upset because Sugar Land was kept intact except for the Avalon subdivision, which was carved out from HD26. She asked that the subdivision, which has a very high Asian population (70- 80%) and had always been in HD26, be put back with the rest of Sugar Land. (E) Donna Klaegez, Burnet County Judge also testified. She testified that commissioners could not come due to the short notice. Chuck Bailey and Chris Chapman testified on behalf of the City of Irving and Las Colinas in Dallas County. Bailey noted that 93a he had gotten short notice of the hearing and had not had a chance to fully prepare. Gabriel Soliz, city council member from Victoria, Texas testified that the hearing was on very short notice. He proposed switching Bee County and Aransas County between HD30 and HD35. (This was not done.) (F) George Korbel, a redistricting expert, testified that there were unnecessary county cuts in the map, including in Hidalgo and Cameron Counties because there was a county cut in Cameron and a cut in Hidalgo, when if the spillover of the two districts were joined, there would only be one county cut. There was a discussion about whether these were county cuts or spillovers. One of the HRC members asked if doing that would push population up and end up diluting the Hispanic population in HD35. Korbel said he had a plug-in fix that would show how it would work. He also testified that the map “shorted” the Hispanic population in the Valley by one district. He noted that a district was taken from Harris County but no new district was created to represent the Hispanic growth there. He noted that minorities are very under-represented in the map and that each step taken in Dallas, Harris, and Hidalgo Counties removed minority representation. He testified that he had not had enough time to really study the proposed map. There was a discussion that SSVR had dropped in HD90 in Tarrant County, and Korbel noted that a drop in SSVR could be problematic. 94a (G) Mini Timmaratu and Jonathan Fong of TAARI testified. Timmaratu testified about Fort Bend County and HD26. She noted that the County has one of the fastest growing Asian populations in Texas, and that the area had elected Indian-American representatives locally, but that the Asian vote had been diluted. She stated that the Asian VAP was 33.6% in HD26 in the benchmark, and this was reduced to 27.5%, and this was a violation of § 5 of the VRA. She testified that the Asian population was split among the four districts in the area and they should be put together. Fong testified about HD149 and the fact that Alief is split into multiple districts. Fong testified that Asian voters were becoming more active and able to elect their candidates of choice. (H) Celeste Villarreal testified on behalf of the Mexican American Bar Association (MABA) in opposition to Plan H113 as infringing on the Hispanic community’s vote. She supported the Task Force/MALDEF Plan H115. Fidel Acevedo testified on behalf of LULAC in favor of Plan H115 and in opposition to Plan H113. Marcelo Tafoya testified on behalf of Hispanics Organized for Political Empowerment (HOPE) against Plan H113 because it did not offer equal Latino electoral opportunity and raised VRA concerns. He said Latinos would lose a district when they should gain districts. He said they leaned in favor of Plan H115. 95a Jose Chavez testified on behalf of himself that Plan H113 did not reflect the population changes. He noted that Tarrant County had seen tremendous minority population growth. He stated that HD90, 93, 95, and 96 were performing minority districts, and although there was a new seat added to Tarrant County, there were only three performing districts—HD90, 95, and 101—in Tarrant County in Plan H113. He said HD90 and HD95 were packed by taking voters from HD96 (which had a combined minority population over 50%). He said they packed three districts and cracked two to end up with only the three performing districts. (I) A member of the HRC noted that HD90 had 40% SSVR, which was down from the benchmark, but it only concerned him somewhat because the district would probably perform given its consistent performance. He also noted that it seemed inconsistent with the growth pattern for the SSVR to decrease. (J) Villarreal stated that they would not vote out the bill, and proposed amendments should be presented when Solomons was present and the committee would vote on the bill. Villarreal then withdrew the committee substitute to H.B. 150 and left H.B. 150 as pending business. D-116 at 102-04. He announced on behalf of Solomons that the next committee meeting would be Tuesday the 19th, that amendments were due by 10:30 a.m. Monday (the next day), 96a and that the mapdrawers were working on a committee substitute to address concerns raised thus far by MALDEF, Alvarado, and Veasey, within certain constraints, and encouraged members to get amendments in as soon as possible so that they could be considered. He stated that the map would not go to the floor for another week. 131. Veasey testified that there was insufficient time for meaningful public input given the short timeframe and the hearing being on Palm Sunday. TrJ17. Rep. Turner also testified that it was very difficult for people to quickly change their plans and get to the hearings. Tr798. He stated that the only explanation given for the short time-frame was that they “were operating on an expedited schedule” even though there were five or six weeks left in the session. Tr798. 132. On April 17 at 9:54 p.m., Hanna sent an email to Denise Davis writing, “I think it would be a mistake to put on floor this week. First, not even sure you can technically do it w. a committee vote on Tuesday [April 19]. Need a calendar rule for prefiling which we would also need to do on Tuesday with a weds 10 am filing deadline for a Thursday vote. They are going to have a new committee substitute out on Tuesday which no one outside the committee chair has seen. Saying you have less than 24 hours to file amendments to this is not good. If anything the lawyers need to review the [committee substitute] to see what legal issues there might be. Rushing the committee report makes errors more possible and thus a [point of order]. I don’t think the members will be happy with having this short of a schedule on something that is very important to them. And while we can fix it in the senate, it would be better to not go there. Please wait until the week after 97a Easter. I know you want to clear this but rushing this is a mistake.” D-261. 133. On Monday April 18, Plan H134, Solomons’ second statewide House proposal was made public. D- 317. Proposed amendments to districts 32, 34, 42, and 80 in Plan H134 (made public as Plan H135) were also made public. D-317. 134. On Tuesday, April 19, the HRC held a formal meeting to consider the House plan. PL-220 (notice given on House floor); US-272B (announced on House floor); PL-207 (minutes); D-116 at 106; D-12. No public testimony was taken. Solomons laid out H.B. 150 as pending business. (A) This meeting was held in the Agriculture museum, which does not have audio and video broadcasting capabilities. PL-207 (minutes); D-12; Tr1464 (Interiano); Tr1574 (Solomons). Accordingly, there is no archived video footage of this meeting. Interiano testified that this was because the House was considering an important education bill on the floor, so the decision was made to have the room be as close as possible to the floor for members to be able to go back and forth. Tr1464. Solomons also testified that the Agriculture Museum is right down the stairs from the House floor and is commonly used for formal hearings because of its convenience to the floor. Tr1573. (B) Rep. Alonzo (Hispanic, Democrat) offered Plan H115 as a statewide committee substitute, and it failed (Ayes: Villarreal, Alonzo, Alvarado, Veasey; Nays: Solomons, Aycock, 98a Branch, Eissler, Geren, Harless, Hilderbran, Hunter, Keffer, Madden, Peña, Phillips). D- 12. Rep. Veasey (African-American, Demo- crat) offered statewide committee substitute Plan H130, and it failed (same votes). D-12. (C) Solomons offered his new committee substitute plan, Plan H134. He said the map increased the number of SSVR districts to 30, that HD40 and HD41 would “flip,” and that the substitute addressed the urban and rural retrogression of the original bill. D-12. It was adopted (Nays: Alonzo, Alvarado, Harless, Veasey; Ayes: Solomons, Villarreal, Aycock, Branch, Eissler, Geren, Hunter, Keffer, Madden, Peña, Phillips; Hilderbran did not vote). Solomons then offered a Tarrant County amendment (Plan H151) to Plan H134, which moved a small area from HD99 to HD97. The amendment was adopted by a unanimous vote of those present (Pickett was absent). D-116 at 108. (D) Veasey offered an amendment (Plan H119) to Plan H134 that would amend the Hidalgo County districts HD31, 36, 39, 40, and 41, but it failed (Ayes: Villarreal, Alonzo, Alvarado, Veasey). Veasey offered two amendments, Plan H120 (Tarrant County) and Plan H123 (Travis County), to Plan H134, but they were withdrawn without objection. (E) Rep. Branch (Anglo, Republican) of Dallas County offered an amendment, Plan H126, and a perfecting amendment Plan H148, to Plan H134 affecting HD100, 103, and 108 99a in Dallas County. D-12. The amendments were adopted. (F) Solomons offered his amendment (Plan H135) to Plan H134, affecting HD80/Webb County and HD32 and HD34 in Nueces County, that was adopted (Nay: Veasey, all else present voting Aye). D-12. (G) No amendments were adopted at this hearing that would have created an additional minority opportunity district. TrJ18 (Veasey). Solomons directed the staff to incorporate the amendments into a committee substitute, and the substitute was adopted without objection. Solomons moved that H.B. 150, as substituted, be reported favorably to the full house, and the motion passed 11 (Solomons, Aycock, Branch, Eissler, Geren, Harless, Hunter, Keffer, Madden, Peña, Phillips) to 5 (Villarreal, Alonzo, Alvarado, Hilderbran, Veasey). Peña (Hispanic, Republican) was the only minority legislator to vote for the plan. C.S.H.B.150 came out of committee as Plan H153. TrJ1939, TrJ1946 (Bruce). Interiano testified that he did not look at election performance of the new districts in terms of the OAG reports until about the time that Plan H153 was voted out of committee. TrA10. 135. On April 19 at 5:08 p.m., Hanna emailed Jennifer Welch and copied Interiano and Bruce with a draft C.S.H.B. 150 calendar rule. D-329. Hanna wrote, “Per Denise 8 pm Thursday but this can easily be altered. New language added so amendments can’t 100a result in other than 150 districts in the bill. No giving Harris 25 without taking away one somewhere else.” 136. On April 20, there was a debate on the floor concerning the proposed calendar rule. TrJ1204 (Hanna); D-190 (transcript) (the Fed. R. Evid. 106 objection is overruled). The original calendar rule made by the Calendars Committee required amendments to be filed by Thursday, April 21 at 8 p.m. David Hanna drafted the original amendment to the calendar rule, which required amendments to be filed by April 22. TrJ1204 (Hanna). (A) Rep. Hunter moved for a calendar rule that would make amendments due by 5 p.m. on Friday, April 22, which was Good Friday. Rep. Turner and Rep. Dutton, both African- American Democrats, objected and wanted additional time to submit amendments. Turner stated that he was “concerned about the limited amount of input that we have had in the drafting of these legislative districts.” He said there were concerns about retrogression in the African-American districts and they were only being given a short period of time to respond with proposed amendments, and that he would be in church on Good Friday. Turner continued, “Chairman Hunter, as chair of the Calendars Committee, I mean, I want to impress on you and the Texas House the abbreviated and expedited timeline under which we are operating. This is a bill that was laid out a week ago Wednesday, there were committee hearings on Friday and Sunday of this past week, a bill that was voted out on Tuesday morning, and now we 101a are being asked to provide amendments to this bill by Friday 5 p.m., with Friday being Good Friday, and with many of us going to participate in Good Friday ceremonies. And I will tell you in the strongest of terms that the Legislative Black Caucus has some serious concerns about the retrogression in HB 150, about the impact, the adverse impact that this bill has on our districts, and deeply concerned about the lack of input that we have had in the total process in the drafting of these bills, that we believe that a calendar rule Friday at 5 o’clock on Good Friday is simply—goes counter to an open and fair process.” (B) Dutton moved to amend the calendar rule to have the deadline for amendments be 5 p.m. on Monday April 25 rather than Friday. D-190 at 13-14. Hunter moved to table, stating that an earlier deadline would give members more time to review proposed amendments because it was going to the floor on Tuesday, and he was not willing to move that deadline. Id. at 14-16. Dutton then argued that the rushed process put a huge target on the plan for a legal challenge based on the process. After further debate and after legislative counsel agreed to be available over Easter weekend to draft amendments for the members, the deadline for filing amendments was moved to 5 p.m. Monday, and the calendar date of the House bill was changed from Tuesday the 26th to Wednesday the 27th. TrJ1947 (Bruce); TrJ1205 (Hanna). This provided additional time for amendments to be filed, 102a and an additional day before floor debate. TrJ1205 (Hanna). Rep. Branch (Anglo, Republican) noted that there was reason to hurry because if they did not pass a bill it would go to the LRB. D-190 at 20. 137. Sometime around April 20, David Hanna wrote a third retrogression memo analyzing Plan H153. US-338; D-123; TrJ1161 (Hanna). Because the SSVR in HD90 and HD148 had been raised, Hanna no longer raised retrogression concerns for those districts. With regard to HD148 in Houston, whereas Hanna identified the SSVR in the second memo as 38.6% and suggested bringing it up, it was now 49% total SSVR, and Hanna wrote, “Assuming all districts perform to allow Hispanic voters [to] elect candidates of choice, the equalization of all of these districts at the 50% SSVR level should not present retrogression issues.” In addition, Hanna removed the language about considering whether a fifth majority Hispanic district could be drawn in Harris County and whether it would be required by § 2 of the VRA. But this memo raised some of the same retrogression issues—Nueces County, HD35, Hidalgo, Dallas County—that still were not resolved. TrJ1163 (Hanna). Regarding Nueces County, he again noted that the County Line Rule would have to yield to the VRA if it could be shown that retrogression could be avoided by splitting the County, but this approach should be pursued only if it would clearly contribute to total Hispanic voting strength statewide. As to HD35, Hanna still recommended further election analysis to measure the performance of the benchmark and proposed districts. With regard to Dallas County, Hanna still noted possible retrogression issues with the two Hispanic districts HD103 and HD104 and suggested remedying

103a the decline in SSVR in HD103. In the Hidalgo County area, the district numbers for HD40 and HD41 were swapped, but the retrogression analysis did not change. TrJ1196 (Hanna). Hanna still noted that Peña’s district dropped by almost 6% in SSVR while the adjoining districts all had substantially higher SSVR, and there was a risk the DOJ would find the change retrogressive through packing Hispanics in other districts. Hanna recommended restoring the district’s SSVR. Hanna did not perform an offset analysis to see if any retrogression could be made up elsewhere, nor did he perform any election analysis regarding HD40 and HD41. TrJ1196 (Hanna). Hanna said his concerns were based on objections that the State received in 1991 and 2001 to similar districts in Cameron County, where there were “the same sort of phenomena of a lower district in terms of Spanish surname voter registration surrounded by two higher districts.” TrJ1196- 97. Hanna felt more certain about his retrogression concerns in this area because of the prior objections. TrJ1197 (Hanna). Regarding El Paso, Hanna still concluded, “The minor decreases in the SSVR in Dis- tricts 77 and 78 will not likely present retrogression issues.” D-123. Hanna did not do a final offset analysis in his third memo or afterward to see if the number of minority seats had decreased. TrJ1195,TrJ1218 (Hanna). Hanna shared his memos with Bruce, Interiano, and Downton. TrJ1177 (Hanna). None of the Hanna memos were shared with minority legislators. TrJ1957 (Bruce). 138. On April 21, David Hanna sent an email to Interiano and Downton with the subject “CSHB 150 benchmarks.” US-126 12 ; D-328; PL-1619; Quesada-

12 The foundation, authentication, hearsay, and relevance objections are overruled. 104a 154; Quesada-265; TrJ1199-1200, TrJ1172-73 (Hanna); TrJ1614 (Interiano). Hanna wrote, “I did some comparisons adding up the total majority Hispanic districts and came up with this: PLAN SSVR SSVR HCVAP total Reg. non-suspense midpoint over 50% over 50% % over 50 H100 29 29 30 H153 26 30 30 H115 33 34 35

Plan H115 was the MALDEF proposal. Hanna continued, “In two of the three possible benchmark indicators, CSHB [Plan H153] meets or exceeds the benchmark in 50% districts. Only in the total registration SSVR number does it fall short. This is because you drew 4 districts just over 50% using RedAppl, which has only the non-suspense SSVR on it. With total reg. they dropped just below 50, though Dist 90 went down to 47.9% I listened again to Luis’s testimony from last Friday and he said the MALDEF plan had 34 districts with an SSVR over 50%. This corresponds to the non-suspense count. He also said they had 35 HCVAP districts. So at least as of last Friday, with the mark MALDEF set, you should be OK. Of course they (or someone) may eventually look at total SSVR and try to make a deal about it. I think it will be very difficult and moderately disruptive to try to get all four of your districts that are just under 50% SSVR on the total reg. to be over 50. Dist 90 is already the smallest district in the state and so I’m guessing you put everything in there that you could. You might be able to get Dist. 104 over 50%. I bet you can get one of the two Houston Districts (145, 148) 105a over 50 but not the other. In 2001 we used only the non-suspense SSVR numbers on our system. No one complained, but there really is no factual basis for using total reg. over non-suspense reg. in calculating SSVR that we know of. The best measure might be somewhere in between. Call me with any questions.” 139. Hanna said this was a § 5 analysis but he conceded that it could alert Downton and Interiano that Plan H153 was potentially vulnerable under § 2 because Plan H115 and other minority-proposed plans had more minority districts; he also noted it was based purely on demographic information and he never tested for election performance. TrJ1174-76 (Hanna). Hanna was not asked to do a separate § 2 analysis. TrJ1176 (Hanna). Hanna said the memo relayed the conclusion that, at least considering the metric used by DOJ in the 2001 objection letter concerning retrogression under § 5, “we were probably okay.” TrJ1200. Hanna agreed that two of the districts that were newly over 50% were already performing and counted as ability districts in the benchmark plan (HD90 and HD148), so they did not necessarily indicate an increase in ability-to-elect districts, and including them in the count could arguably mask the loss of an ability district like HD33. TrJ1216-17 (“I think you could make that argument.”). This was the only offset- type analysis Hanna remembered performing. Fur- ther, Hanna understood § 5 to require performing districts, but did not analyze the performance of the various districts in the benchmark or proposed plan. TrJ1211 (Hanna). Although mapdrawers had been using only SSVR data up to that point, Hanna used both SSVR and HCVAP in his analysis because HCVAP data was now available. TrJ1199-1200 (Hanna). HCVAP numbers were not available in RedAppl, but 106a TLC could run reports when requested. TrJ1615 (Interiano). 140. Interiano and Downton took Hanna’s email as “the green light” to go to the floor with the plan. TrJ1532 (Interiano); TrJ2139 (Downton). Interiano concluded that “we were taking a fair and legal map to the floor” because they were meeting the benchmarks of SSVR and HCVAP. TrJ1614-15 (Interiano). Hanna did not intend to advise and did not advise that the map complied with the VRA. TrJ1164 (Hanna). 141. On April 22, Stacy Napier from OAG sent an email to Bruce and Downton with the subject “Racially Polarized Voting Summaries.” US-190/US-190A; TrJ1957-58 (Bruce recalls receiving such RPVAs from OAG). Attached were racially polarized voting summaries for certain districts in Plan H153. She wrote, “Attached are the summaries we discussed providing yesterday. From our number crunchers: Given the low turnout and Hispanic citizenship in districts 137 and 144, we didn’t feel comfortable identifying a candidate of choice from the regressions, but 137 seems to have been strengthened for minority candidates whereas 144 appears to have been weakened. We can discuss this on Monday if you have questions.” Bruce forwarded the email with the attachments to Downton, Interiano, and Solomons that same day. The attachment contained a spreadsheet consisting of ten elections (the OAG 10) for various districts. US-190/190A; TrJ1019 (Solomons). The spreadsheet contained a column identifying the minority preferred candidate and provided an analysis of districts in Plan H153. US- 190; TrJ1020-22 (Solomons). It showed that in Peña’s district HD41, the Hispanic candidate of choice would have prevailed in 7/10 elections under Plan H100, while in Plan H153 it was 5/10 elections. It also had 107a summaries for HD40, HD78, HD137, HD139, HD140, HD141, HD144, HD147, and HD148. Interiano knew from the OAG RPVAs that Hispanic performance in HD41, HD35, and HD117 decreased in Plan H153, but he felt it could be offset with increased performance in other districts. TrA9-10, TrA34, TrA37 (Interiano); US-3. The analysis also showed HD148 staying at 10/10. D-141. 142. On April 22, Lamar Smith emailed Opiela asking whether the Solomons House map or the Nixon House map (H212, a competing map being proposed by Rep. Nixon) was better with regard to districts “below 55.” Perez-124; TrJ2083 (Downton). Opiela then forwarded the message to Interiano and Denise Davis, asking if they had “the data on # VRA and # of GOP under 55% comparison between Solomons/Nixon State house Plans for Lamar.” Perez-124. Interiano responded, “Below 55 of what number . . . don’t share this with ANYONE but here is a comparison of the SSVR districts. Nixon map drops it by the current benchmark of 29 to 27. As far as Republican seats, he is overreaching. He drew two seats in Travis and both were lost by McCain and Perry.” Id. Opiela responded, “Lamar meant below 55 McCain or Perry. Is there an analysis already done between the two using those benchmarks?” US-128. Interiano answered, “I do, but all of that is in the office at work.” Id. Opiela replied, “I’ll wait on it and tell Lamar we’ll get it for him on Monday. Didn’t look forward to running stats on 300 districts.” Id. Interiano then wrote, “Tell him Monday or Tuesday . . . but what does he want to do with it? That’s not something that we are sharing with anyone.” Id. 143. On April 26, Rogene Calvert of TAARI, Nina Perales of MALDEF, Dr. D.Z. Colfield of the NAACP 108a Houston, and Mustaafa Tameez of TAARI wrote a letter to Solomons complaining about the loss of HD149 in Harris County. US-267; Quesada-342 (the objections to this exhibit are overruled). 144. On April 27, Nina Perales of MALDEF wrote a letter to Solomons to provide information on Plan H153. PL-227; D-15. She wrote that Plan H153 was retrogressive and raised serious concerns under the VRA. She noted that the DOJ retrogression analysis from 2001 did not turn solely on the number of districts in a plan that contained a majority of SSVR, and the ultimate evaluation turned on all the relevant circumstances, including the ability to elect a candidate of choice, opportunity to participate in the political process, and the feasibility of creating a nonret- rogressive plan. She noted that Plan H153 eliminated a Latino opportunity district (HD33) in Nueces County. She further wrote, “Plan H153 also raises the Latino population in two existing Hispanic voting age majority districts, Districts 90 and 148. Our analysis of primary and general elections has led us to conclude that Districts 90 and 148 in the benchmark plan are districts in which Latino voters currently possess the ability to elect their candidates of choice. Adding Latino voters to Districts 90 and 148 does raise the SSVR but does not create new Latino opportunity districts that can offset the loss of District 33.” She continued, “Thus Plan H153 reduces the number of districts in which Latinos can elect their candidate of choice.” PL-227. 145. On Wednesday April 27, Plan H153 was brought to the House floor for second reading/debate. TrJ1947 (Bruce); D-13; D-190_00087 (proposed maps and reports associated with amendments). The House 109a debated second reading all day on the 27th. TrJ1947 (Bruce); D-190 at 630/S702. (A) Solomons laid out Plan H153 and stated that not all members had provided proposed maps, but that the map represented the input provided to “our offices, and committee office, for the members of this house and members of the public throughout the interim hearings throughout the state, and three hearings during this session.” D-13 at S99. He stated that it was his intent to limit the pairings as much as possible and to give each member the opportunity to win their district such that no Republicans and Dem- ocrats were paired together. Id. The pairings were as follows: Reps. Flynn and Cain; Ritter and Hamilton; Landtroop and Perry; Chisum and Hardcastle; Driver and Burkett; Harper-Brown and Anderson; Scott and Torres (all Republicans); and Hochberg and Vo (both Democrats). He stated that the pairings resulted in eight new districts that had to be reincorporated in the full map, and they were predominantly anchored in the suburban counties across the state that experienced the largest growth, including Montgomery, Brazos, McLennan, Collins, Fort Bend, Wise, Tarrant, Denton, and Williamson Counties. Id. at S100. He stated, “Recognizing the growth of the Hispanic population in this state, we tried to create a new Hispanic opportunity district and strengthen other Hispanic districts. First, statewide, the map increases the number of districts with Span- ish surname voter registration exceeding 50 110a percent from 29 to 30. Specifically we increased the SSR-SSVR- in District 90 in Tarrant County and District 148 in Harris County. Both of these changes were made at the request of MALDEF. Second, we created a new strong Hispanic seat in Nueces County, and as a whole, Nueces has an SSVR of 49 percent. So, it would not have been possible to create two Hispanic seats within Nueces County. Third, we created a new coalition minority opportunity district in Tarrant County [HD101].” Id. He said the map was a “work in progress.” Id. (B) Rep. Martinez (Hispanic, Democrat) noted that although Solomons had said it would be a member-driven map, a lot of amendments from members, especially from the Valley, were not taken into account or adopted. D- 13 at S100. Martinez asked who Gerardo Interiano was, and Solomons responded that he was working out of Speaker Straus’s office, but at Solomons’ direction. Id. at S100-101. Martinez commented that he, Rep. Gonzales, and Rep. Muñoz had testified against the Valley configuration and also raised a point of order on the basis that the legislators were not included in the minutes. Id. at S102. (C) Rep. Dutton (African American, Democrat) asked if the HRC had taken the census data as it was given or whether they had made any adjustments to it. Solomons responded that no adjustments were made and that prisoners were counted at their place of incarceration. D-13 at S103-105. Solomons 111a admitted that counting them at their pre- incarceration address could have changed Harris County (so it would have 25 districts). Id. (D) Solomons stated that the courts, not him, would have to decide that the VRA should preempt Texas’s County Line Rule. D-13 at S106. (E) Rep. Farrar (Hispanic, Democrat) asked why the map did not increase the number of effective minority opportunity districts given that 89% of the growth was minority. D-13 at S111. Solomons said they increased the number of minority districts from 29 to 30 and “beefed up” a couple of districts, including Farrar’s, for SSVR. Id. He stated that one of the primary indexes they used was SSVR. Id. Farrar pointed out that they did not need to increase the SSVR in HD148 because it was already electing Hispanic representatives. Id. (F) In response to questioning from Rep. Turner, Solomons stated that the benchmark had two districts with 50% BVAP and 11 with 40% BVAP, and the proposed map had two 50% BVAP and 12 40% BVAP districts. D-13 at S113-S114. Solomons stated that staff reached out to the NAACP but did not receive any specific feedback from them. Id. at S114. Turner stated that no one had reached out to him as chair of the Texas Legislative Black Caucus for any input at all. Id. at S115. Turner also complained that the “timeline has been very, very abbreviated,” to which Solomons 112a responded that he had been “pretty available.” Id. (G) Rep. Veasey pointed out that Anglo population had decreased by 198,000 in Dallas County and all the growth was minority, that although Anglos were only about 33% of Dallas County they still controlled 68% of districts, and that the numbers did not seem to be adding up. D13 at S116. Solomons replied that it was a “member driven” map for Dallas County, that the committee did not see the need to create new districts, and that primarily the growth “was in a lot of existing protected districts” and not “in the relevant areas.” Id. (H) Regarding Tarrant County, Veasey (who represented HD95 in Tarrant County) questioned why HD93 was retrogressed. Solomons stated it was a drop-in map from the delegation. D13 at S116. Veasey conceded that they had submitted an agreed delegation map, but noted that his and Rep. Burnam’s districts (HD90) had been changed without their approval, but with the approval of people in Anglo districts. Id. at S117. Rep. Geren responded that HD93 was a minority district, but that became HD101, and that MALDEF asked them to increase the HVAP of HD90. Id. Veasey responded that legally that meant nothing because HD90 was already a Hispanic opportunity district, and HD90 now came far into his district HD95. Id. Geren replied that “it has more Hispanic opportunity now than what it was.” Id. at S118. Solomons 113a said the SSVR numbers of HD90 and HD148 “were increased slightly . . . because we were advised basically we needed to do that just for additional protection to make sure that they were where they needed to be.” Id. Solomons said the only reason they increased the SSVR of HD90 was because MALDEF requested it and its SSVR was only at 45% before. Id. at S119-S120. (I) With regard to Nueces County, Solomons said that the County Line Rule required two districts in Nueces County because it was at 2.03% of ideal district size, and the overall population in Nueces was 49% SSVR such that “it is impossible to draw two Hispanic majority seats within Nueces.” D-13 at S120-121. He said that they decided to draw only one strongly Hispanic district to allow the Hispanic community in Nueces to elect a representative of their choice, and all three members of the Nueces County delegation agreed to that. Id. at S121. Rep. Martinez Fischer stated that, had they cut a county line, they could have had two minority opportunity districts in Nueces County, and that Plan H153 had 17 cuts, so one could have been in Nueces County. Id. at S121. Solomons responded that there was only one county cut and everything else was spillovers. Id. (J) Martinez Fischer asked why HD31 was now placed into Hidalgo County and noted that it arguably “results in the prevention of 114a drawing an additional [Hispanic opportunity] district in South Texas.” D-13 at S122. (K) Rep. Walle (Hispanic Democrat), who represented HD140 in Harris County, noted that Harris County had grown at the same rate as the state but was losing one seat, going from 25 to 24. D-13 at S124. He also noted that Anglo population declined in Harris County (by 82,000) while minority population increased (Hispanic by 551,789; African-American by 134,564; Asian by 76,827), yet no new minority opportunity districts were created, an effective coalition district, HD149, was eliminated, and the HVAP in HD137 (Hochberg) was reduced by 4.4%. Id. Solomons stated that the Texas Constitution required them to round down and pointed out that in 2011 the House passed a map with 24 districts in Harris County, and it was the LRB that issued a map with 25 seats. Solomons stated that Reps. Coleman, Dutton, Farrar, Hochberg, Thompson, and Turner had all voted for a 24-district map. Id. Solomons asserted that the 25-district aspect of the LRB plan was never challenged in court, but they did not want to take the legal risk. Id. at S125. He stated that their legal team did not think HD137 or HD149 were protected districts, though he recognized that there was disagreement on that. Id. Rep. Hochberg noted that he had previously testified before the HRC on April 17 (when Solomons was absent) that 2001 was not the first time the 115a county had been rounded up to 25 and previously no one had said they could not. Id. at S126. Solomons responded that there had not been a court challenge and the safer course was to round down. Id. (L) Rep. Yvonne Davis (African American, Dem- ocrat) inquired about what the committee considered to be retrogression. D-13 at S127. Solomons said that the “people behind the scenes” looked at a lot and he was not clear on everything the lawyers looked at. But he said, “I do know that they analyzed approximately ten elections for the elected offices which we are affecting and determined whether the candidate of choice of the minority majority would have won with the change[d] precincts of the new district. And I can’t tell you nor suggest that anybody on the committee tell you a real detailed explanation of what they do with precincts which were not originally in the district and where that candidate was not an option. But there are several programs that run in the retrogression analysis and that’s what they do.” He said that when the retrogression analysis indicates there might be a problem, “we try to make some adjustments.” Id. at S128. He said, “it’s hard to tell exactly other than they give you and they say in the last ten elections here’s what happened. They have a modeled computer pro- gram, and they are trying to get as close as they can to ensure that they can comply with the Voting Rights Act.” Id. 116a (M) Rep. Menendez (Hispanic, Democrat) offered Amendment 1 (Plan H160), which changed some lines between his district and Rep. Castro’s district (HD124 and HD125 in Bexar County) that involved no people and was agreed between them and was acceptable to the author. It was adopted. D-190 at 88-89, S729; D-13 at S130. (N) Rep. Farias (Hispanic, Democrat) put forth Amendment 2 (Plan H182) regarding his district and Rep. Garza’s district in Bexar County (HD117 and HD118). D-13 at S130; D-190 at 90-91. His amendment would put parts of four precincts that were moved into Garza’s district (Whispering Winds) back into Farias’s district. D-13 at S130-131. Farias stated that these were poor, rural areas that he would like back, and that the amendment would keep Garza’s district at 50.1% minority, which was a “big concern” for Garza. Farias stated that Garza had told him that he would not oppose the amendment, but would live with the will of the House. Id. at S132-133. However, Garza did not get up to say it should be left to the will of the House. Aliseda asked how it would affect Republican performance, and Farias stated, “We didn’t look at republican numbers because the last conversation we had was that his big concern that he did have a minority district, and he wanted to keep it without changing the numbers. So we kept it at 50.1 percent that he had. He said he would be happy with that.” Id. at S132. Farias further stated 117a that the areas in question likely would not support Garza because Garza is a Republi- can and they were Democrat areas and Garza was getting military precincts that have “some republican votes.” Id. Farias was a third-term member while Garza was a freshman, and having represented the area, Farias did not think those areas would vote for Garza. (O) Solomons stated, “The proposed District 117 was drawn to bring together rural and suburban communities and interests in south and west Bexar County. The amendment as proposed creates a jagged and awkward line that divides these communities and creates an island along I-37 that connects to the west of the district by only a tiny strip of land. And a proposed map cleanly follows the proposed map as it exists today, follows the Medina River, a natural barrier that divides the rural and urban counties–communities in south Bexar County. The amendment reduces the compactness of both 117 and 118. It increases number of split precincts in District 117 from one to seven and in District 118 from nine to fifteen, and increases the number of split VTD’s in District 111 from one to seven and in District 118 from 10 to 15. It does have a negative impact on District 117[‘s] republican numbers. It is a district that—it goes from—let’s see—pretty much—that’s pretty much it. I guess the will of the house—I’m going to make a motion to table but it really will become a matter for the will of the House, I suppose, but I’m going to oppose it 118a because of the reasons stated. It’s trying to lay out what I think the amendment does in connection with compactness and increases the split of precincts and increases the number of split VTDs.” D-13 at S133-134. Amendment No. 2 was tabled. (P) Rep. Anderson (Anglo, Republican) laid out Amendments No. 3 (Plan H220) and 4 (Plan H247) relating to Dallas County that would pair Rep. Sheets and Hartnett and Driver and Button instead of Anderson and Harper-Brown. D-190 at 92-95, S733; D-13 at S134-S135. Anderson stated that he believed the district as currently drawn was fair and legal but he “believe[d] there’s another way to keep the core districts and communities of interest together. It keeps Mesquite, Garland, Richardson, Carrollton, Addison, Irving, and Grand Prairie predominantly intact. It also reflects the diverse neighborhoods that are located within the city of Dallas.” Id. at S134. Rep. Martinez Fischer stated that the proposed map might be retrogressive because it dropped HD104’s SSVR from 58.3% to 45.5% and it did not create any new opportunity districts. Id. at S138, S141. Republicans Hartnett, Jackson, Branch, and Harper- Brown did not support the amendment. Solomons moved to table, and it was tabled. Id. at S142-143. Rep. Driver (Anglo, Repub- lican) complained that he had no input into Dallas County and that his community of interest was “cut up.” Id. at S137, S141. Anderson complained that the proposed 119a map “dramatically chang[ed] representation throughout all of Dallas County. Id. at S142. (Q) Rep. Harper-Brown (Anglo, Republican) proposed Amendments 5 and 6 (Plan H219) relating to Dallas County. Harper-Brown asserted that the amendment would keep more of Irving in HD105 and allow HD115 to encompass Carrollton and Farmers Branch and protect the community of interest of these cities and improve compactness. Rep. Giddings (African American, Demo- crat) complained that it destroyed the core of HD109, a minority opportunity district. D-13 at S144. Rep. Anderson complained that it divided Grand Prairie, “the 15th largest city in Texas,” into five pieces in Dallas County. Id. at S145. Rep. Yvonne Davis asserted that it changed HD111 from a majority-African-American district to majority Hispanic and opposed the amendment. Id. at S146. Rep. Anchia complained that it reduced the SSVR of his district and opposed the amendment. Id. at S148. Solomons stated, “it breaks up Grand Prairie far more than what the original map does. It redraws the west side of Dallas County. It reduces Representative Anchia’s SSVR numbers problematically simply but not enough for Leg Council to have an over- concern . . . .” Id. at S148. In discussing retrogression of an African-American district (HD110), Solomons said that one of the primary things to look at is VAP. Id. at S152. Rep. Turner opposed the amendment on behalf of the Legislative Black Caucus 120a “because we view it as outright regression.” Id. at S155. Rep. Driver (Anglo, Republican) took issue with Solomons’ statement that “the Dallas delegation was all inclusive in the drawing of the Dallas County map,” stating, “It was not all inclusive. That’s a false–and it was false information he was given about it.” Id. at S157. It was tabled after debate. Id. at S158; D-190 at 96-97. (R) Rep. Ritter (Anglo, Republican) and Rep. Deshotel (African-American, Democrat) proposed Amendment No. 7 (Plan H145) that moved one block from Ritter’s district to Deshotel’s. It was agreed and acceptable to the author and was adopted. D-13 at S158. (S) Rep. Rodriguez (Hispanic, Democrat) proposed Amendment Nos. 8 and 9 (Plan H123) to add Precinct 101 to his district from Rep. Dukes’s district (HD46), and put Precinct 222 back in her district in Travis County. Dukes said they were attempting to prevent any retrogression of African-American voters in HD46 and prevent retrogression of His- panic voters in HD51 and they came to a painless solution. D-13 at S159. Solomons left it to the will of the House, and the amendment was adopted. D-13 at S160; D-190 at 104-105. (T) Rep. Veasey (African American, Democrat) proposed Amendment No. 10 (Plan H120) for Tarrant County that would have created a new minority opportunity district (HD96) in Tarrant County. D-13 at S160; D-190 at 106-107. Veasey stated that Plan H153 dropped a minority opportunity district 121a from the benchmark Tarrant County (from four to three) and packed his district and Burnam’s unnecessarily, despite the large minority growth. Veasey further noted that Tarrant County had agreed on a map but his (HD95) and Burnam’s (HD90) districts, the two most minority districts, had been changed. D-13 at S161. Veasey stated, “There’s a big line going through the middle of my district. Representative Geren picked up the Como community, and I think everybody over there knows Charlie [Geren] and I think they are more comfortable being represented by Lon [Burnam]. But if you look at the major changes that took place from after the point in which we signed off, the big change that has happened in the districts were represented by the minorities, and we didn’t approve them.” Id. at S163. Burnam and Veasey stated that they did not approve of the whole Tarrant County map, only their districts, which were then changed. Id. Rep. Truitt (Anglo, Republi- can) stated that Veasey’s proposed map split communities with common interests right down the middle. Id. at S162. Burnam said Lake Como had always been in HD90 but was now split. Id. at S163. Veasey said that HD96 (represented by Zedler, Anglo, Republican) was one of the fastest growing minority areas in the county and had “a shot to decide who gets elected in that district” but did not anymore. Id. at S164. Rep. Geren (HD99, Anglo, Republican) moved to table the amendment, stating that it split communities of interest and dropped the 122a HVAP of HD90 below 50%, and that there were only three minority districts in the benchmark. Id. at S164-165. Burnam stated that he was elected in a district that was 47.2% SSVR, and although the map raised it over 50%, it was done by making his district 8,200 plus people below the ideal population and by taking out the Lake Como community, which is an overwhelming minority-majority community, and putting it into Geren’s district. Id. at S165. Burnam and Veasey argued that the map was retrogressive, that HD90 was already effective, that the map did not represent Tarrant County, and that HD96 was an effective coalition district when there was good turnout, and it had been cracked to assure re- election of a white representative. Id. at S167. Geren pointed out that changes were also made to districts represented by Nash, Patrick, and Shelton, but Veasey said they were only “slight changes.” Id. The amendment was tabled. (U) Rep. Burnam (Anglo, Democrat) laid out Amendment No. 11 (Plan H203) and a perfecting amendment, Amendment No. 12 (Plan H236) for Tarrant County. D-13 at S168; D-190 at 108-111. He said he thought they had an agreed-to plan acceptable for HD90, which was established by a federal judge in 1978. D-13 at S168. He said it was ugly, but it fixed HD90 and put Lake Como back. It was only 49.1% SSVR, but the 50% number was arbitrary, and the DOJ had rejected using a fixed demographic percentage in favor of a functional analysis. 123a Id. Burnam stated that the predominantly African-American district in Tarrant County was packed, that his district represents “the artificial nuance of just two or three percentage points in the registered His- panic voters, when, in fact, the proposed map takes out over 10 percent of the voting population in this district, which includes the minority constituency in the Lake Como community. It rips it out and puts it in a district in far northwest Tarrant County in Azle. And while the people in Azle are wonderful folks they just don’t have very much in common with my inner-city Fort Worth constituency, the Lake Como community. I’m extremely concerned that in a district that is largely a no growth district that you would put me at such a low number. It undermines the one person, one vote.” Id. at S169. Solomons stated that there were problems with the amendment because it reduced the SSVR of HD90 below 50% (to 49.1%) and that MALDEF had testified that it needed to be above 50%. Burnam asked him if he had heard Veasey talking about the letter from representatives of the Latino community stating that the 50% criteria was not particularly important in this instance, and Solomons stated that the letter did not repudiate and take back the MALDEF testimony on the record. Solomons stated that they were trying to be consistent and conservative about legal risk. Id. at S171. Burnam also asked him about the DOJ guidelines and stated that “it looks a little hypocritical to people 124a that are looking at it closely because if you were really following the MALDEF recommendation you would create five new His- panic districts not just call Representative Farrar’s district and my district new His- panic districts. Those two communities in Harris County and Tarrant County already vote for who they want to vote for, and those minority communities don’t want to have their districts ripped up.” Id. at S172. Solomons moved to table. Veasey and Burnam reiterated that Plan H153 was “all a contrivance to look like we’re not in retrogression” and that there would be a lawsuit if the map was not changed. The amendment was tabled. Id. at S173. (V) Rep. Walle (Hispanic, Democrat) from Harris County laid out Amendment No. 13 (Plan H172) that would “create a community of interest of Sugar Land and for the Asian American residents of Fort Bend County.” D-13 at S174. He stated the committee map seemed to be convoluted and violated the city boundaries; its split precincts fractured the City of Sugar Land and diluted Asian voting power in Fort Bend County. Id. The proposed amendment would put over 30% “other” (Asian Americans) into HD26, and it would be a coalition district with an opportunity for Asians to reflect the Asian population growth in Fort Bend County. Rep. C. Howard (HD26, Anglo, Republican) noted that the members of the delegation had already agreed on a plan, and asserted that he adequately represented the Asian community. Walle argued that Asians 125a needed representation, and that it was wrong to eliminate Vo’s district HD149 in Harris County. Id. at S176. The amendment was tabled. (W) Rep. Hunter (Anglo, Republican) laid out Amendment No. 14 (Plan H178) that he said moved a precinct in Nueces County and was approved by the committee. It was adopted. (X) Rep. Morrison (Anglo, Republican) proposed Amendment No. 15 (Plan H180) that would move DeWitt County back into his district (HD30) and move Karnes County into Rep. Kleinschmidt’s district (HD17). It was agreed and acceptable to the author, and it passed. D-13 at S178-S179. (Y) Rep. T. King (Anglo, Democrat) proposed Amendment No. 16 (Plan H161) that affected LaSalle County and Jim Hogg County. D-13 at S179. The amendment would move La Salle County out of HD35 and into HD80, and move Jim Hogg County into HD35, represented by Aliseda. Rep. Raymond (His- panic, Democrat) then proposed Amend- ment 17 (Plan H242) to Amendment 16, which involved no people, and simply moved some landmarks. It was adopted. Id. Rep. Aliseda stated that he was not happy with the amendment (Plan H161) because he was a freshman who would likely draw an opponent and it would reduce his Republican numbers. Id. at S180. Aliseda said it reduced his HVAP but increased his SSVR “a little bit more.” Id. at S183. Rep. 126a Villarreal responded that SSVR is the relevant population to get a sense of ability to elect. Id. at S183. Villarreal argued that SSVR in HD35 dropped, and the amendment would bring it up. Id. Aliseda said he did not oppose the increase in SSVR but “the fact that it changes my district into a more democrat district.” Id. at S184. Aliseda moved to table, and it was tabled. Id. at S186. (Z) Rep. V. Gonzales, Rep. Martinez, and Rep. Muñoz (Hispanic Democrats) presented Amendment No. 18 (Plan H187) relating to Hidalgo County. D-190 at 123-24; D-13 at S186. Gonzales discussed the major changes that had been made to Hidalgo County districts, including that her district and Rep. Peña’s district had essentially been swapped and each had 1.5% or less of their prior districts. D-13 at S186. She stated that she would have to start all over with member-constituent relationships. Id. at S190. She noted that Martinez’s district was also changed (he stated that he had only 72% of his district) and Muñoz had only 57.2% of his prior district. She also noted that the HD40 and HD41 district numbers had been changed and that they tried to make Peña’s district a Republican district. She noted that 14 VTDs are split, despite Solomons objecting to Farias’s proposed amendment based on an increase in split VTDs. She also asserted that the map packed and cracked Hispanics, complained that the committee map did not create a new district in the Rio Grande Valley 127a despite the population growth, and noted that Peña’s district would be one of the smallest in the State. Id. at S186-S187. Gonzales stated that the proposed amendment would put “us basically back to where we were” by adjusting the population and preserving communities of interest, without splitting any VTDs. Id. at S188. She also stated that she believed there was retrogression and the plan would not be precleared. Id. at S189. Rep. Aliseda (Hispanic, Republican) noted that MALDEF’s representative Figueroa had previously testified that the Hidalgo County map would not violate the VRA, and Gonzales responded that Figueroa had been concerned with § 2, not § 5, and he was not aware of the issues. Id. at S190. Aliseda asked if the proposed amendment would affect Peña’s Republican numbers. Rep. Oliveira (Hispanic, Demo- crat) stated that combining Cameron and Hidalgo County could create seven districts in the Valley and create a brand new Latino opportunity district while keeping the other districts. Id. at S191-S192. Oliveira also criticized the over-reliance on the County Line Rule, and he and Burnam asserted that the Legislature was being inconsistent on its use of the rule and its insistence on 50% SSVR. Id. at S192-193. Rep. Guillen noted that Gonzales’s amendment did not create a seventh district, but Oliveira noted that those maps were laid out in other plans. Id. at S193. Oliveira stated that Guillen was protected by being given Hidalgo County population that he did not need to keep his 128a minority opportunity district, but “that population was sacrificed for your benefit and perhaps for Representative Peña,” and Guillen stated that he was not sure “about the logic behind” his assertion that it was drawn for Guillen’s benefit. Id. at S193- S194. Rep. Martinez asked who drew the lines for the Hidalgo County map, and neither Solomons nor Villarreal knew. Id. at S194-195. Martinez noted that someone other than the delegation had to have given input because the map was not drawn the way he, Muñoz, and Gonzales asked, even though it was supposed to be member- driven. Id. at S195. Villarreal noted that he opposed the map and voted against it coming out of committee because it failed to create a district that combined Cameron and Hidalgo population. Id. at S195. Peña stated, “I did not draw this map. But there are certain things I do know and certain things I did ask for. I said one, please don’t pair any of my colleagues. . . . The other thing I said was if there’s going to be spillover I would rather have an experienced member come into the valley than a freshman. And so Representative Guillen, who you know, is somebody who is respected, senior member, so that was my suggestion. Those are the two suggestions I made.” Id. at S195-S196. He argued that he should be in a conservative district and Gonzales in a Democratic district, and the map was fair. Id. at S196. Peña stated he

129a would love for there to be another Valley district but that the County Line Rule would be followed unless a court said it could be ignored. Id. Martinez asked Peña if he had input on the lines being drawn in Hidalgo County. Peña stated that he had “some input,” including that members not be paired, that the conservative member get a conservative district, that other members represent the areas they have, and that a senior member like Rep. Guillen come in if there was spillover. Id. at S197. When asked if he had input on his own district, Peña said, “No. When I met with the members what I said to you was, look I expect that the republicans are going to maximize the conservative seats. That’s what I told you, you can recall that. And I said I will not draw this map because one, I did not want to be involved. And two, that I didn’t want to be involved in pairing or being involved in affecting my neighbor’s districts.” Id. at S198. Martinez asked him “So, the drawing of your lines in your district you didn’t have any input in?” and Peña said “No, I never even bothered to learn the redappl.” Id. Rep. Castro pointed out that the objection to Farias’s amendment in Bexar County was that it created a jagged line, yet HD41 was full of jagged lines. Peña stated “the entire map has that sort of thing” and that he did not draw it. Id. at S199. Peña moved to table, and the amendment was tabled. Id. at S203. 130a (AA) Rep. Gonzales offered another Hidalgo County amendment, Amendment No. 19 (Plan H162) that would try to keep the core of the districts the same but create more of a Republican district for Peña as more of a compromise than the previous amendment. D-13 at S203. Solomons told members to “vote their consciences.” Id. It was tabled. Id. (BB) Rep. Smithee (Anglo, Republican) offered Amendment No. 20 (Plan H154) concerning West Texas. D-13 at S203. He stated two significant changes would be avoiding the unnecessary pairing of Landtroop and Perry and eliminating part of “that district [HD88] that has drawn so much attention, that linear district that goes pretty much to the width of Texas, over 350 miles.” Id. He also stated it would preserve communities of interest and make the districts more geographically compact. Id. at S204. Rep. Perry (Anglo, Republican) offered Amendment No. 21 (H206) to Amendment No. 20 to “clean up some inter-district stuff.” It would affect HD84 in Lubbock County by removing a “little finger” that was created before and was politically motivated but no longer needed. Id. at S206. Rep. Frullo (Anglo, Republican) opposed it. Id. at S204. Rep. Martinez Fischer advised that members of MALDEF should “sit this one out” because they should not vote for maps that they may not ultimately agree on in the bigger picture. Id. at S205. Amend- ment No. 21 was tabled. Rep. Gallego 131a (Hispanic, Democrat) then offered Amend- ment No. 22 (H250) to Amendment No. 20 to add Loving County to HD74, and it was acceptable to the author and adopted. Id. at S206. Amendment No. 20, as amended, was adopted. Id. (CC) Rep. Woolley (Anglo, Republican) laid out Amendment No. 23 to Harris County (Plan H191), and Rep. Coleman (African American, Democrat) laid out an amendment to that amendment, Amendment No. 24 (Plan H271), reflecting changes made through negotiations among Harris County members. D-13 at S206; D-190 at 134-37; D-53(3) (map). The floor debate was stopped for several hours to allow Democrats to work with Interiano and some Republican members of the delegation to make changes to the Harris County map. Tr159 (Martinez Fischer); Tr933-34 (Downton). Woolley and Coleman presented the agreed amendments. Woolley noted that “we all worked together in the back room.” D-13 at S207. Coleman assured Woolley that his district, HD147, which went from 39.2% to 38.2% BVAP, was still an African- American opportunity district (though not an African-American majority district) and was not retrogressed. Id. Woolley moved for adoption of Plan H191 as amended by Coleman’s amendment and it was adopted, as amended. Id. at S208. (DD Amendment No. 25 (Plan H159) to Harris County to swap a block between his and 132a Rep. Farrar’s district that was agreed, and it was adopted. D-13 at S208. Rep. Alonzo (Hispanic, Democrat) laid ut Amendment No. 26 (Plan H164), “the MALDEF amendment.” Alonzo said it would add opportunity districts, including two in West Texas and one in Lubbock. D-13 at S208. After a debate about the County Line Rule, the amendment was withdrawn because it inadvert- ently paired two Dallas members. Id. at S209. Alonzo stated that he did not intend for the plan to have a legal effect. Id. at S210. (EE) Rep. Martinez Fischer (Hispanic, Demo- crat) laid out Amendment No. 27 (Plan H195) that presented a MALC statewide plan, and perfecting Amendment No. 28 (Plan H269) for Dallas and Kaufman Counties, which was adopted to restore Johnson’s BVAP to benchmark. D-13 at S210; D-190 at190-244. Martinez Fischer and Rep. Gallego discussed the lack of Latino districts in West Texas, Odessa, Amarillo, Lubbock, Midland, and the panhandle. D-13 at S211. Martinez Fischer stated that Latinos represent 40% of the West Texas population and their numbers had maintained growth, while Anglo population had declined, yet no growth was reflected in the proposed plan. Id. He stated that the map demonstrated that minority opportunity districts could be created if there was a will, but that it was not done and instead other districts were 133a packed to avoid creating new districts. Id. at S212. His map created additional minority districts in West Texas (2), South Texas, DFW, and Harris County. Id. at S213. Solomons stated that Martinez Fisher’s amendment unnecessarily violated the County Line Rule numerous times and moved to table. Martinez Fischer and Solomons debated the County Line Rule. Solomons reiterated his position that he would follow it, and argued that HD31 in Hidalgo County was spillover and not a county cut. The House voted not to allow Martinez Fischer additional time to debate the amendment, which Martinez Fischer took issue with given that they had given the Harris County delegation three hours to work on their County map and it was not his fault that the hour was late. Id. at S216. Martinez Fischer’s amendment (Plan H195, amended by H269) was tabled. Id. at S217. (FF) Martinez Fischer laid out another statewide amendment (Amendment No. 29) (Plan H196). Rep. V. Gonzales emphasized the Latino growth and the need for a seventh Valley seat (HD72) by combining Cameron and Hidalgo Counties. D-13 at S217-S218. Citing the County Line Rule and prior debate, Solomons moved to table, and it was tabled. Id. at S218. (GG) Martinez Fischer proposed statewide Amendment No. 30 (Plan H197) to create additional minority seats in West Texas. 134a Reps. Gallego and Alonzo discussed that the plan was trying to address the increase in Hispanic population in West Texas, even though the overall population decline caused a loss of two seats there. D- 13 at S218-S219. Solomons asserted that the maps violated the County Line Rule and moved to table. Id. at S219. It was tabled. (HH) Martinez Fischer proposed an amendment (Amend. 31) (Plan H198) regarding the Coastal Bend. D-13 at S219. Rep. Raymond complained that minority voters in Nueces County were diluted and that the Nueces County configuration violated § 5. He stated that African- American neighborhoods had been placed into HD32, diluting their strength, Hispanic precincts were split and diluted, and a Hispanic opportunity district had been eliminated. He advocated for the MALC plan. Citing the County Line Rule, Solomons moved to table, and the amendment was tabled. Id. at S220. (II) Martinez Fischer then set forth Amend- ment No. 32 (H199) focusing on Dallas County and Tarrant County to increase the number of opportunity districts from three to five. He stated that Plan H153 violated § 2 of the VRA by limiting the creation of minority opportunity districts. A statement of legal issues regarding the DFW metroplex and the April 27, 2011 letter from MALDEF stating that Plan H153 was retrogressive were made part of 135a the record. D-13 at S224. The statement of legal issues noted that Plan H153 substantially overpopulated Latino majority districts HD103 and HD104 in Dallas and across the state. It concluded that the overpopulation of Latino majority districts in Dallas County served to limit Latino influence in Dallas County house districts. Id. at S221-222. It argued that use of the County Line Rule fenced apart minority voters in Dallas County and Tarrant County, and that Plan H153 packed minority districts, and failed to create additional minority opportunity districts, while alternative plans created up to two by unpacking districts and avoiding the County Line Rule. Id. at S222. The amendment was tabled. Id. at S230. (JJ) Martinez Fischer then laid out Amend- ment No. 33 (Plan H200) focusing on Har- ris County. Rep. Walle asserted that Plan H153 violated the VRA and Texas Consti- tution in Harris County, and the MALC plan remedied the issues. Walle stated that reducing the number of districts to 24 diluted minority representation in Harris County, and that Plan H153 packed existing Latino opportunity districts and failed to create any new Latino opportunity districts. D-13 at S230. He said the MALC plan created two more minority districts than Plan H153, and one more than the benchmark. He and Rep. Hochberg asserted that HD137 was a protected district. Id. at S231. Hochberg 136a noted that the LRB had stated his district HD137 was a fifth majority-HVAP district when submitting it for preclearance, but that it had not been treated as a protected district. Id. at S232. Hochberg pointed out that there was a VTD split that actually split an apartment complex (this was later fixed) and that the map cut the Sharpstown community of interest. Id. Walle complained about the loss of Rep. Vo’s district and the dilution of the Asian vote. Id. at S234. Vo and Walle discussed that HD149 was a majority- minority district, with about 20% Asians, 20% Hispanics, and 20% African Ameri- cans. Vo noted that witnesses had testified in support of HD149 and hundreds of emails had been sent to the members of the HRC. Id. at S233. Solomons stated that he believed the map was legal and moved to table. Id. at S234. Solomons stated that HD137 and HD149 were not protected districts. Id. Walle stated, “[W]e were not informed that a 24 map was going to be submitted. Okay. That map was submitted to you without any of the other democratic members from Harris County being informed of such map. And that’s the issue that many of us from Harris County have is that we were not informed that those maps had been submitted, signed off on, a 24 map. When we were trying to negotiate to get a 25 map. And for us, we take that very personally because one, you are eliminat- 137a ing Representative Hubert Vo and pairing him with Representative Hochberg. And at the same time packing the other Latino - other majority minority districts. So for us, the process wasn’t clean for us.” Id. at S235. Walle also asked why every minority-majority district in Harris County was overpopulated, which Solomons did not answer. The amendment was tabled. Id. at S236. (KK) Martinez Fischer then laid out Amend- ment No. 34 (Plan H20113) that he intended to withdraw to continue the discussion on the County Line Rule. He stated that Plan H153 had 17 spillovers or county breaks, and his amendment had a statewide map with the same number of county breaks, while still increasing minority opportunity. He stated that Solomons had 30 SSVR-majority districts, while this one had 32; that Solomons had 34 HVAP- majority seats and this plan had 37; that Solomons had two BVAP-majority districts and this had three; that Solomons had 53 B+H VAP-majority districts and this had 59; and Solomons had 30 HCVAP-majority districts and this had 31. D-13 at S236. Rep. Aliseda (Hispanic, Republican) objected that all Martinez Fischer’s maps had “messed with [his]

13 Plans H195, H196, H197, H198, H199, and H200 introduced by Martinez Fischer were essentially the same map introduced as different amendments, and Martinez Fischer used each one to focus on a particular area. Plan H201 was a different map from the others. D-13 at S237. 138a district” and also noted that this turned his district, HD35, into a predominantly Anglo district. Martinez Fischer responded that this was the result of strictly following the County Line Rule rather than growing minority opportunities. Id. at S237. He stated that the map was not about partisan politics because “if minorities are truly representing those minority opportunity districts then we don’t have to worry about politics.” Id. Aliseda disagreed and noted that all the maps seemed to turn Republican districts into Demo- cratic districts, although it also increased minority representation. Id. at S238. The amendment was withdrawn. Id. (LL) Rep. Turner (African American, Demo- crat) laid out Amendment No. 35 (Plan H202, co-authored by Rep. Y. Davis (African American, Democrat)) and a perfecting amendment (Plan H227) for Dallas County, presenting a statewide map on behalf of the Texas Legislative Black Caucus (“LBC”). D-13 at S238. Turner stated, “Other than for today you have not heard much conversation from African Americans with respect to this entire redistricting process. There has been very little input provided to this committee process as it relates to African Americans.” Id. He stated that his proposal was not political and that 16 of 17 members of the LBC (which has two Republican members) had signed it. He stated that Plan H153 negatively and adversely impacted African-American 139a representation, that it was retrogressive, and that input had been very limited. He stated that the plan had 63 minority opportunity seats (including 13 BVAP >38%; and 13 B+HVAP>50%) and one Asian plurality district in Fort Bend County (HD26). Id. at S239. He said minority representation was enhanced in HD84 in Lubbock. Solomons moved to table, stating that “he still splits counties” and “I think there are some legal issues in connection with how he splits communities of interest and also the county lines.” Id. at S240. Turner stated that “it seemed as though the [redistricting] discussion was very much one-sided where African- Americans were not included in any appreciable sense in the discussion as it relates to the redistricting process.” Id. at S244. Solomons moved to table, stating that the plan “splits a number of counties unnecessarily,” and it was tabled. Id. at S242. Solomons again stated that he would not break the County Line Rule unless Texas changed the law or a federal court said he should do something different. Id. (MM) Rep. Alvarado (Hispanic, Democrat) set forth Amendment No. 37 (Plan H226) that had 25 Harris County districts and “maximizes minority majority opportunity districts.” D-13 at S245. She stated that it had 57 total districts where voters can elect a candidate of choice and created five new coalition districts. Alvarado pointed out that Harris County had been 140a rounded up for “the last couple of decades.” Id. She also stated that a good portion of the Harris County delegation did not have input into the first map that was drawn at 24, and that it was not member- driven. Id. Solomons moved to table, stating that the committee map was legal and increased the number of minority- majority districts to 30, that there were two BVAP over 50% districts, and that other maps broke county lines. Id. at S246. He also stated that the proposed map created 25 districts in violation of the Texas Constitution. The amendment was tabled. Id. at S247. (NN) Rep. Coleman (African American, Demo- crat) set forth Amendment No. 38, a statewide amendment (Plan H232), that would create a number of opportunity and coalition districts (59 in total). Coleman said it strengthened three existing but weak Latino districts (HD35, HD78, HD137), in contrast to how Plan H153 treated them. He asked that his written points be put into the record. They state that Plan H232 has 59 minority opportunity districts, 36 Latino opportunity districts, 13 African-American districts, and 10 coalition districts. Solomons stated that the plan was problematic because it reduced the number of Hispanic-majority seats based on SSVR from 29 to 27, reduced the number of BVAP-majority districts from 2 to 1, and created 9 pairings. The amendment was tabled. D-13 at S250. 141a (OO) Rep. Giddings (African American, Demo- crat) noted that Dallas County members had submitted maps and they were not accepted. Giddings stated that the proposed committee map was problematic because it did not create influence districts and there was racial disparity in terms of African Americans having less representation than could have been achieved, and the communities of interest were not preserved in minority districts. D-13 at S253. Rep. Villarreal stated that he was going to vote against C.S.H.B. 150 because it did not do enough for minority opportunity. Id. at S254. Rep. Hochberg also stated that he would be voting against the map because of its treatment of Harris County (even if he were not paired). Id. at S255. (PP) The House passed C.S.H.B. 150, as amended, to engrossment, by a record vote of 92 Yeas, 52 Nays, 5 present not voting. (stipulated); MALC-77 (record vote). Democrats Eiland (Anglo, Democrat) and Guillen (Hispanic, Democrat) voted for the plan. The plan as passed out of second reading was Plan H276. 146. On April 28, C.S.H.B. 150 (the House plan) went back to the House floor for third reading. TrJ1947 (Bruce); D-190 at S846. Solomons noted that there were a few agreed-to amendments “that help make the bill better, and they’ll be acceptable to the author, I believe.” D-190 at S846. (A) Rep. Creighton (Anglo, Republican) laid out Amendment No. 1 (Plan H268) to change a 142a precinct between himself and Rep. Eissler (Anglo, Republican), and it was adopted. Id. (B) Rep. Raymond (Hispanic, Democrat) laid out two amendments (Plans H277 and H279) that put the Laredo Energy Arena and the Casa Blanca State Park into his district and out of Rep. King’s district, and they were adopted. Id. One amendment was later withdrawn. (C) Rep. Wayne Smith (Anglo, Republican) proposed Plan H281 with amendments to several districts in Harris County (affecting districts 126, 127, 128, 139, 141, 142, 143, and 150), and it was adopted. D-190 at S846; D-53(4)(map). This Amendment was the result of further negotiations with Rep. Senfronia Thompson (African American, Democrat) and some Republican and Dem- ocratic members of the Harris County delegation. (D) Rep. Allen (African-American, Democrat) offered an amendment (Plan H282) to keep her office in her district, with only changes to HD131 and H146 in Harris County, and it was acceptable to the author. It was adopted. D-190 at S848; D-53(5)(map). (E) Rep. Veasey (African-American, Democrat) asked Solomons about the proposed congressional map from Congressman Lamar Smith. He said he had read a newspaper article stating that Congressman Smith had delivered a congressional map to Solomons and to the House, and he was wondering when Solomons was planning on 143a unveiling that particular map or when the House or the HRC would take it up. Solomons replied, “I have no idea yet. I actually haven’t read the article and haven’t seen the map.” He also stated, “I haven’t even looked at a congressional map yet. And the earliest that anyone would be able to do anything, including me and my staff; would probably be next week at the earliest.” D-190 at S847. (F) Rep. Y. Davis (African American, Demo- crat) stated that she “wanted to go back and revisit some discussion we had last night as it relates to the Legislative Black Caucus Plan 202.” D-190 at S848. Davis stated that Solomons had opposed the plan was because of County Line Rule violations. Solomons asserted that it split Nueces County four ways, Cameron County three ways, and unnecessarily split Upshur, Wilson, Goliad, and Victoria Counties. He also stated that it split both Henderson and Hill Counties, and while they had to split one of those counties, they did not need to split both. Davis asserted that Bartlett v. Strickland held that a district had to have a 50% voting age population to overcome the state constitutional ban and “therefore, you could [split county lines] if, in fact, you were over the 50 percent level. And those counties like Cameron County, Willacy, and Hidalgo, and Cameron actually exceeded the 50 percent guideline, which allows you then to be able to split those.” D-190 at S848-849. 144a (G) Rep. Cain (Anglo, Republican) spoke against the map, claiming that it fractured core districts and broke up communities of interest, and that the members from East Texas had virtually no input into their districts. He said, “My district alone was divided into four districts. Another district, relatively compact, was stretched over hundreds of miles. Communities of interest were fractured.” D-190 at S849-850. He proposed the Nixon map as an alternative. (H) Solomons then thanked the members for their participation and moved passage. H.B. 150, as amended, was passed by Record Vote No. 596: 92 Yeas, 54 Nays, 3 Present, not voting. Yeas—Aliseda; Anderson, C.; Anderson, R.; Aycock; Bohac; Bonnen; Branch; Brown; Burkett; Button; Callegari; Carter; Chisum; Cook; Craddick; Creighton; Crownover; Darby; Davis, J.; Davis, S.; Eiland; Eissler; Elkins; Fletcher; Frullo; Garza; Geren; Gonzales, L.; Gooden; Guillen; Hamilton; Hancock; Hardcastle; Harless; Hartnett; Hilderbran; Hopson; Howard, C.; Huberty; Hunter; Isaac; Jackson; Keffer; King, P.; King, S.; Kleinschmidt; Kolkhorst; Kuempel; Landtroop; Larson; Laubenberg; Legler; Lewis; Lozano; Lyne; Madden; Margo; Miller, D.; Miller, S.; Morrison; Murphy; Nash; Orr; Otto; Parker; Patrick; Peña; Perry; Phillips; Pitts; Price; Riddle; Ritter; Schwertner; Scott; Sheets; Sheffield; Shelton; Simpson; Smith, T.; Smith, W.; Smithee; Solomons; Taylor, L.; Taylor, V.; Torres; Truitt; Weber; White; Woolley; Workman; Zerwas. Nays—Allen; 145a Alonzo; Alvarado; Anchia; Berman; Burnam; Cain; Castro; Christian; Coleman; Davis, Y.; Deshotel; Driver; Dukes; Dutton; Farias; Farrar; Flynn; Gallego; Giddings; Gonzales, V.; Gonzalez; Gutierrez; Harper-Brown; Hernandez Luna; Hochberg; Hughes; Johnson; King, T.; Lavender; Lucio; Mallory Caraway; Marquez; Martinez; Martinez Fischer; McClendon; Menendez; Miles; Muñoz; Naishtat; Oliveira; Paxton; Pickett; Quintanilla; Raymond; Reynolds; Rodriguez; Thompson; Turner; Veasey; Villarreal; Vo; Walle; Zedler. Present, not voting—Mr. Speaker; Howard, D.; Strama. Absent— Beck (would have voted yes). Lozano stated he was shown voting yes but he intended to vote no because he believed that outside of his district, other district lines violated the U.S. Constitution, Texas Constitution, and VRA. Lozano testified that he voted for Plan H283 because his constituents wanted him to vote for a map that did not pair them with Nueces, but he eventually changed his vote because of threats by Martinez Fischer. TrJ1793-1796 (Lozano). The vote split largely along partisan lines. All Dem- ocrats except Eiland (Anglo) and Guillen (Hispanic) voted against the plan. African- American (Carter and White), Hispanic (Aliseda, Garza, L. Gonzales, Peña, and Torres), and Asian (Button) Republicans voted for the plan, and 10 Republicans voted against. 147. Hanna gave his opinion that the County Line Rule must yield to the VRA, and this advice was included in the redistricting book given to the 146a Legislature. TrJ1208-1209 (Hanna). Hanna never advised anyone during the redistricting process that the VRA would have to yield to the Texas County Line Rule; he would have said the opposite. TrJ1209 (Hanna). No one in the Legislature asked him to analyze whether the County Line Rule prevented them from creating additional minority opportunity districts. TrJ1210 (Hanna). Despite this contrary advice, Solomons was clear during the process that he “was going to stick with the county line rule because that was the law at the time.” TrJ1011 (Solomons). He made his intention to follow the County Line Rule clear to everyone. TrJ1012 (Solomons). If a Latino district could not be drawn within the constraints of the Texas Constitution it was not done. Tr1447 (Interiano). The risk of not abiding by the County Line Rule was that the map would be challenged in state court and be ruled illegal before it would ever get to federal court, and then jurisdiction would have gone to the LRB. TrJ1201 (Hanna). 148. On April 29, the OAG produced a report titled “Hispanic Population Profile” that showed each district in Plan H283 with the total and % HVAP according to the 2010 Census, the total and % HCVAP based on both “TLC CVAP” and “OAG CVAP” and the total and non-suspense SSVR. US-43. 149. Hanna never saw an election analysis that would suggest that Plan H283, the final enacted plan, was not retrogressive. TrJ1164 (Hanna). Hanna never told anyone that he believed that Plan H283 complied with the VRA. TrJ1164, TrA1508 (Hanna). May 2011 147a 150. In May 2011, Congressman Lamar Smith and the Republican congressional delegation were continuing to advocate to legislative leadership for a new VRA district in the DFW area in the congressional plan. TrA246 (Seliger); PL-311. 151. On May 4, Bruce emailed Interiano and Tony Essalih, Chief of Staff for Congressman Culberson, in response to his questioning about when the HRC might meet. PL-311. Bruce wrote that no schedule had been agreed to and “discussions are ongoing between the House and the Senate.” She said she would let all members of the congressional delegation who had inquired know of Solomons’ intent once a schedule was concrete. 152. On May 6, the SRC met to consider H.B. 150, the House map. D-16; D-598 (transcript); US-272C. (A) Seliger stated that it was customary for the Senate to have a hearing on the House map as required by Senate rules, but not to amend the House map. D-598 at 6. (There had been some discussion between the Senate and House redistricting committee staff about whether the Senate Committee had to have a public hearing before the Senate could consider the bill on the floor, and the Senate Committee decided to have a hearing.) Seliger said they would honor the House map “as is,” and the House would honor the Senate map, as was customary. (B) Witnesses testified. Rep. Harper-Brown (Anglo, Republican), who represented HD105 in Dallas County, testified against the bill as it related to her district, offering 148a an amendment that she claimed did not violate the VRA as the proposed map did and limited the splitting of communities of interest. She testified that HD105 was being changed significantly, and that Irving was being split into three different districts, which destroyed its ability to maintain effective representation by one House member. D-598 at 23. She questioned whether the House plan would be approved by the DOJ. Chris Wallace of Irving testified in favor of having HD105 be wholly within the city limits of Irving. Id. at 17. Irving Mayor Pro Tem Rick Stopfer and Mayor Herbert Gears also asked that the City of Irving and its corporate tax base be kept whole in a district. Id. at 17. Sandra Crenshaw from Dallas testified that the proposed map split cities and African-American communities of interest. She also complained, as a precinct chairman, about the numerous split precincts, noting that it costs money and confuses voters. Id. at 22. She asked at a minimum that some of the split precincts be cleaned up. The bill was left pending. Id. at 24. 153. On May 9, 2011, the Perez v. Perry, 5:11-cv-360 (W.D. Tex.), and MALC v. Texas, 5:11-cv-361 (W.D. Tex.) lawsuits were filed. Shannon Perez is a Bexar County voter and Harold Dutton, Jr. is a Harris County voter and legislator. Plaintiff Mexican American Legislative Caucus (“MALC”) is a Latino legislative caucus. MALC is a non-profit organization established to serve the members of the Texas House of Represent- 149a atives and their staff in matters of interest to the Mex- ican American community of Texas, in order to form a strong and cohesive voice on those matters in the legislative process, including redistricting. Many of MALC’s members are elected from and represent con- stituencies in majority-Latino districts, and many of its members are Latino. (stipulated) 154. On May 13, the SRC met to consider the Senate plan and the House plan. D-18; D-117 at 58; D- 600 (transcript). The House plan was voted favorably out of committee without discussion or debate. D-18; D-117 at 61; D-600 at 53-54. 155. On May 14, 2011, Opiela emailed Denise Davis stating that Congressman Smith’s office had not seen the congressional proposal, asking if Speaker staff had seen it, and asking if they could set up a meeting to work together on a map. D-609; US-746. At this time, Smith was still advocating for his proposed plan. Quesada-68 at 45. On Monday May 16, 2011, Opiela emailed Denise Davis to ask if she thought it would be beneficial to send a copy of their proposal memo to Solomons and Seliger to remind them “of the various aspects and rationale behind the delegation’s proposal.” US-492, US-745; D-610. 156. On May 17, 2011, Opiela forwarded to Interiano an email sent by Lamar Smith to “Texas Col- leagues.” US-749; D-611; PL-1025. The email provided “talking points” to use to encourage senators to adopt the delegation proposal, including that “the delegation proposal allocates the new districts in a three Repub- lican, one Democratic allocation, with a 25-11 Republican/ Democratic split for the entire map,” the delegation proposal “keeps all incumbent Republican congressmen at 55% McCain or higher, with the exception of Cong. Canseco, who is in a VRA District;” “the 150a delegation proposal does not retrogress existing minority districts, and minimizes the risk of a successful court challenge by adding two new Hispanic districts (DFW and Doggett)”; “the delegation proposal adds a new DFW VRA district for two reasons - (1) to minimize the possibility of a successful Section 2 claim since there are over a million Voting Age Hispanics in DFW, and they do not have either an effective coalition or Hispanic majority district and (2) politically, if a new Democratic district is not created in DFW, the long-term prospects of Cong. Sessions and to some extent, Cong. Marchant’s districts as Republican majority districts becomes unlikely in the out years of this decade.” The email continues, “The delegation proposal adds a new Central Texas VRA district in lieu of the current Anglo-majority district occupied by Cong. Doggett to reflect the growth of Hispanics in the Austin-San Antonio corridor. This also is to minimize the risk of a successful Section 2 claim since possible plaintiffs (LULAC/MALDEF) have also demonstrated the viability of a coalition district there. This also provides us with the legal basis (creating a new Hispanic opportunity district) for not significantly increasing the Spanish Surname Registered Voter percentage in CD23 which would endanger Cong. Canseco, since the only source for those voters would be on the heavily Democratic south side of San Antonio). It improves the proportionality of the entire map, both demograph- ically and politically.” 157. Seliger concluded that drawing an HCVAP- majority district in DFW would lead to an unusually shaped district. TrA246 (Seliger). It was possible to draw a minority-majority district that combined His- panics and African Americans in DFW (a coalition district), as Congressman Smith proposed, but Seliger concluded that such a district was not required by the 151a VRA. TrA247 (Seliger). Seliger deferred to Solomons and his staff in DFW because Solomons is from that area. TrA246 (Seliger). Downton was unable to draw a DFW district that was over 50% HCVAP using the 2005-2009 ACS data available through the TLC, so he concluded that a Latino opportunity district was not required to be drawn there. Tr906, Tr981 (Downton); Downton 8-12-11 depo at 68. Solomons also concluded that no coalition district was required in DFW based on advice he received. Because they believed a minority-coalition district in DFW would be a Demo- crat district, mapdrawers and Republican legislative leaders refused to draw such a district unless they determined that it was compelled by the VRA. Tr907 (Downton). 158. On May 17, 2011, the Senate passed H.B. 150 (Plan H283) to third reading by a vote of 22 yeas and 9 nays on second reading. (stipulated). That same day, notice was given of a hearing on May 19 on congressional redistricting. The hearing was canceled the next day. US-611; Quesada-6; PL-213. 159. During this time, both the House and the Sen- ate redistricting committee staff (Downton and Doug Davis) were working on draft congressional maps. TrA1596 (Downton). From the beginning of the process, Downton was drawing a district with Nueces County going to the north, with Solomons’ approval. TrA1773 (Downton). The reconfiguration of CD27 was to help Farenthold get re-elected. TrA229 (Seliger). Seliger believed that CD23 in the benchmark plan was a Latino opportunity district. TrA219-20 (Seliger). His goal at the start of redistricting was to figure out if there was any way “in political terms” to help Canseco hold the seat. TrA222 (Seliger). The Senate redistrict- 152a ers could not figure a way to help Canseco while keeping the district a Latino opportunity district. TrA223 (Seliger). The configuration for CD23 eventually came from the House. TrA224 (Seliger). In addition, by mid- to-late May, Downton was drawing the map without a new Latino opportunity district in the DFW area. TrA1772 (Downton). Downton created only one new Latino opportunity district (CD35) that included parts of Austin and San Antonio. Downton got the idea or this district from a MALDEF proposed map. Toward the end of the regular session, the House and Senate redistricting committee chairmen (Solomons and Seliger) agreed to move forward with the map being drafted by Downton. TrA1596 (Downton). Accordingly, the congressional map was created in the House, and Downton was the primary mapdrawer on the congressional map. TrA1591 (Downton); TrA275, TrA296 (Interiano). Downton relied on CVAP data, as provided by TLC, as a primary metric for drawing the congressional map. He did not consider the lagging nature of the ACS survey data or its effect on the maps. Tr982 (Downton). Interiano also did not consider the accuracy of the ACS data, believing that to be the TLC’s role. Tr1488-89 (Interiano). 160. On May 21, 2011, Seliger moved to suspend the regular order of business to take up for consideration H.B. 150, the House Plan H283, on its third reading and final passage in the Senate. The motion prevailed. Sen. Gallegos asked to be recorded as voting “Nay” on suspension of the regular order of business. The bill was read a third time and was passed with 24 Yeas and 7 Nays. Some Democrat and minority Senators voted for the House plan, following the tradi- tion that the Senate generally will not interfere with passage of the House plan. 153a 161. On May 28, 2011, Downton sent an email to Interiano and Doug Davis stating that he “got the election analysis back from the AG’s office - 10 out of 10 on the slot district [CD35] for electing Hispanic candidates of choice.” US-630 (the hearsay objection is overruled); PL-1658. Interiano responded, “Any guidance on your 23? Have you been able to make any of the changes that we all discussed?” US-630; PL-1659. Downton replied, “Have it over 59% HCVAP, but still at 1/10. There has to be some level of HCVAP where it doesn’t make a difference what the election results are. It is more Hispanic than the other two San Antonio based districts. Changes made to keep the black population together in District 12.” Id. Downton closed, “Anything from the AG on the other issues we discussed with them?” PL-1659. 162. On May 30, Lamar Smith emailed Opiela asking for Downton’s email and asking if it would help Canseco “if I gave him 3k more in bexar (either gop or hispanics) and took edwards co in exchange.” D-612; Quesada-144. Opiela responded, “I don’t think we mess with quico’s district–for your sake and his. His is barely performing (or not depending on your measure) right now; add Rs (which will be Anglos) and you put a neon sign on it telling the court to redraw it. Bring down your numbers and you’ll have a Demo opponent every time. And they won’t be Lainey Melnick.” D-612. Smith responded, “Still want to make offer re edwards. Only 3k. Maybe .1 percent but cld help quico.” D-613. He also said, “Also didn’t realize I had part of guadalupe.” Id. Opiela then responded and copied Interiano: “I didn’t think Lamar had Guadalupe . . . but it’s in the system printout is this a mistake, Gerardo? It’s only 13k people.” Id. Interiano wrote, “I don’t think that it was a mistake. I think this was done in order to make the VRA district work. But I can double check with 154a Ryan tomorrow morning. Did you already pass that information along to Ryan?” Id. 163. The regular session of the 82nd Legislature adjourned (sine die) on May 30, 2011. That same day, the Rodriguez v. Perry, 1:11-cv-00451 (W.D. Tex.), lawsuit was filed. The Rodriguez Plaintiffs are individual voters as well as the City of Austin and Travis County. 164. Governor Rick Perry called a special session on May 30, 2011. The 30-day special session length is in the Constitution and cannot change. TEX. CONST. art. III § 40; TrA1557 (Hanna). Although a special session lasts thirty days, the committee would need to pass a bill in the third week of June because there are House and Senate rules that regulate legislation in a special session; they do not really have the full thirty days. TrA1088 (Hunter). Special sessions are always fast; according to Hanna, the process in 2011 did not seem unusually fast. TrA1557 (Hanna). Time-lines are abbreviated, including that only 24-hour notice is required for a public hearing, instead of five days. TrA1558 (Hanna); D-669. If there had been any proce- dural irregularities, a point of order would have been raised. TrA1558 (Hanna). If the Legislature had not passed the congressional map during the special session, the Governor could have called another special session. TrA1562 (Hanna). If he chose not to, the task would have fallen to the courts. Id. 165. On May 31, Bruce sent Doug Davis (copying Interiano and Downton) an email titled “Hearing.” US- 620 (the hearsay objection is overruled because this is not admitted for its truth). It said, “Congressional Redistricting is going to be added to the call in about 90 minutes. The Chairman [Solomons] is not going to be available all day on Friday and we have a couple other committee members who will not be available 155a on Friday, so are planning on having a hearing on Thursday, June 2, 2011 in the Capitol Auditorium at 10:30 am. This would preclude the necessity of a Joint hearing with the Senate committee. If ya’ll [the SRC] still do Friday, then that would be two days of hearings on the map.” Id. 166. On May 31, 2011, Governor Perry added congressional redistricting to the call for the Legislature’s special session. 167. On May 31, Opiela wrote to Congressman Marchant concerning the Legislature’s progress on the map and describing his district (CD24). D-693. Marchant replied, “Eric, one change, it’s easy, no population. My grand babies go to Hockaday School on forest lane AND Inwood. I have the north side of forest, Pete has the south side. Please go across the street and pluck the campus out of Pete and put in my district. There will be no population. this agreeable and I will ask Burt to do it.” Id. Opiela forwarded Marchant’s request to Downton and Interiano on May 31 stating, “Please see below; if possible.” Id. 168. Seliger and Solomons jointly released their first public congressional redistricting plan (Plan C125) on May 31, 2011. TrA256 (Seliger); US-728 (PLAN log); D-572. The press release announcing Plan C125 noted that it created four new districts, that there were two open Hispanic-majority seats (CD34 and CD35), that it increased the number of Hispanic- majority seats from seven to eight, and that it maintained the three existing Black opportunity seats. D-572. 169. Plan C125 was drawn by Downton. In that plan, Farenthold’s district, CD27, which had been anchored in Nueces County and gone south to 156a Cameron County, was anchored in Nueces County and then spun to the north along the coast and to the northwest into central Texas. Because this district was no longer a Latino opportunity district, CD34 was added as a new Latino opportunity district anchored in Cameron County, extending north to Kleberg County, and wrapping around CD27 to the north. CD33 was added, anchored in Parker County and containing parts of Tarrant County. No new Latino or coalition district was placed in DFW as had been advocated by Congressman Smith, the Governor, and others. Instead, much of the urban and minority populations of Dallas and Tarrant Counties were split apart and connected to suburban districts. CD35 was added as an HCVAP-majority district along the I-35 corridor connecting parts of Travis County/Austin with parts of Bexar County/San Antonio. CD36, which was referred to as the “jumbo-shrimp” or “horseshoe” district because of its shape, was placed north of Houston. CD23, Canseco’s district, was maintained as an HCVAP- majority district, but there were questions about whether it actually performed for Latinos. The other existing HCVAP-majority districts were also maintained. CD30, which was considered an African- American district, was maintained wholly within Dallas County. The two African-American districts in the Houston area, CD18 and CD9, were also maintained. No new African-American districts were added. 170. On May 30 or 31, Hanna asked Archer of the TLC to look at CD23 because Doug Davis had raised some concerns. TrA646 (Archer). Archer did some analyses on CD23 as proposed in Plan C125. He compared the SSVR and election returns of CD23 and CD15, which he considered a “solid Hispanic opportunity district,” and looked at racial bloc voting analyses on CD23. Id. The bloc voting analysis was done through 157a software that TLC had developed in 2000, but it was put “on a back shelf” and only Archer had access to it. TrA649 (Archer). The scatterplots indicated “a pretty high likelihood [of] polarized voting.” TrA651- 52 (Archer). With regard to election results and SSVR, Archer made notes about the SSVR in general elections in CD23 (2002 - 53.2%, 2004 - 54%, 2006 - 53.9%, 2008 - 53.4%, and 2010 - 52.8%). US-609 at 13 (the hearsay objection is overruled because this exhibit is not admitted for its truth). But Archer did not draw a conclusion concerning whether the numbers were consistent with Hispanic electoral opportunity. TrA654 (Archer). He testified that the numbers “make you think at least there is a theoretical competitiveness to that district for Hispanic voters, but you immediately think of, is the polarization of white voters so great that they might outvote a slightly less [cohesive] bloc vote of Hispanics? Is the turnout different? Is the data crummy for the – or at least suspicious? Is it within the margin of error?” Id. Archer noted that Chavez-Thompson, the Latino candidate of choice, was “trounced” by Dewhurst in CD23 in 2010 in the Lt. Governor’s race, and that if he saw that “happening consistently, [he] would be concerned that you don’t have the right electoral activity, the right electoral dynamics for Hispanic-preferred candidates to have an opportunity to be elected.” TrA658. He concluded that, if 2008 were a representative year, this was a competitive district and possibly “a perfectly good, legal district,” but if 2010 were more representative, the district had problems. TrA661 (Archer). Archer wanted to pass the information along but did not draw a conclusion about CD23. TrA675 (Archer). He expressed his concern to Hanna. TrA661 (Archer). He passed on a summary, essentially noting the SSVR and election returns, and said, “you need to look at 158a this closely and decide for yourself what you have got here, because 2010 could indicate a problem.” TrA662 (Archer). He did not discuss his thoughts directly with Downton, but he was confident that Hanna passed them on. Id. 171. On May 31 or June 1, notice was given that there would be a hearing before the HRC on the congressional map (H.B. 4) on June 2. Quesada-6/US- 611 (May 31); D-116_118 (June 1). Notice was also given that there would be a hearing before the SRC on the congressional map (S.B. 4) on June 3. H.B. 4 and S.B. 4 were identical plans (Plan C125). June 2011 172. In June 2011, the TLC was still not able to provide reports showing Spanish Surname voter turnout. TrJ287 (Dyer). 173. On June 1, Interiano sent himself an email titled, “Congressional One Pager” with information on the Solomons-Seliger Congressional Map (Plan C125). US-621 (the hearsay objection is overruled because this is not admitted for its truth). Interiano noted that Ron Paul, Joe Barton, and Bill Flores were “upset Republicans.” He wrote there were four new districts, two new Republican districts (CD33 in Tarrant, CD36 in Harris), and two new Democrat districts (South Texas, Central Texas). Under “Democrat Districts,” he wrote: “New Hispanic VRA district (CD35) from Austin to San Antonio,” and “Hispanic VRA district (CD34) anchored in the Valley. Farenthold’s district became an Anglo-district so the former CD27, now CD34, was strengthened as a Hispanic VRA district.” Interiano described the map as a 3-1 map even though there were two new Democrat districts because 159a Doggett’s district (CD25) was “flipped into a Repub- lican seat.” Regarding “political make-up,” Interiano noted, “All districts, with the exception of CD23 (Canseco), are above 55% McCain. The statewide ORVS [Optimal Republican Voting Strength], according to Baselice, was 58.1% in 2010. 21 Districts are above 58.1%, 4 are below but within .5%, and CD23 is at 53.7% due to it being a VRA district.” Regarding “demographic make-up,” Interiano wrote, “In the current Congressional map, there are six Hispanic VRA districts in South Texas. In the proposed Solomons map, there are 7, an increase of one. With that said, there are going to be two open Hispanic seats on the map. As far as the Black VRA districts, those remain unchanged, a total of 3.” Id. 174. On June 1, Solomons sent a memo to the members of the Texas House. D-649. It stated that he and Seliger had released Plan C125 on the morning of Tuesday, May 31, that the public could access it through District Viewer, and that members could access it through RedAppl or view two paper copies placed in the back hallway. The memo continued, “In order to accomplish our assigned task within the constitutional timeframe for the special session, Senator Seliger and I plan to get to work right away. We have posted back to back hearings for the House and Senate Redistricting Committees this week. The House Committee on Redistricting will meet tomorrow, Thursday, June 2nd, at 10:45 am or upon adjournment in the Capitol Auditorium, while the Senate Committee on Redistricting will meet Friday, June 3rd, at 9:00 am in E1.016. This will provide individuals with an opportunity to testify in both chambers on the proposed map. It is not my intention to vote on the map at this hearing. We will provide 160a notice for a future hearing where the House Commit- tee on Redistricting will vote on the map and any changes which are submitted to the committee.” 175. On June 2, 2011, Opiela emailed Interiano, Downton, and Doug Davis, stating, “Talked to Lamar last night. He’s willing to give up Blanco County to CD25 (or if you need it to Conaway to make the San Saba/Mills swap work so you don’t have to split Burnet) to allow y’all a wider transit through Hays to make it look better, so long as the numbers stay the same. Just wanted to put that out there as an option for y’all.” PL-311; D-615. Opiela also forwarded an email to Downton from Lamar Smith stating, “Hensarling office pct 2220. And neighborhood pct 2223. Let me kno if they will do.” D-694. At 5:02 p.m., Opiela forwarded an email from Andrew Duke with the subject “minor changes.” D-695. He asked if some areas could be swapped between Hensarling and Ses- sions’ districts in Dallas. 176. On June 2, Hanna emailed Doug Davis with the subject “OK to talk to Heath?” US-623 (the hearsay objection is overruled because this is not admitted for its truth); PL-1622. Heath was outside counsel for the SRC. Hanna wrote, “Slight concern on 23; Archer is wondering whether it is really effective in the proposed map. Really no good legal theory to strike it down, but [end of email].” 177. On June 2, Congresswoman Sheila Jackson Lee issued a statement “Regarding Texas Redistrict- ing Proposal Plan C125.” NAACP-608; D-664. She wrote to express her opposition to Plan C125, stating that she was disappointed that it did not reflect the major growth that occurred amongst African Ameri- cans, Hispanics, and Asians in the last decade, and that she believed the plan was “clearly retrogressive” 161a and violated the VRA. She also noted that in CD18, historical neighborhoods are divided and major communities of interest have been carved from the district. She wrote, “I hope this redistricting process will allow more public hearings and an opportunity to fairly protect the rights of minorities to choose a representative of their choosing.” 178. On June 2, the HRC held a public hearing in the Capitol Auditorium on congressional redistricting. PL-217; D-116 at 117; D-20 (minutes); D-601 (transcript, some text missing). Solomons announced that the requirements for committee amendments in the regular session were still effective. D-601 at 3. He said Plan C125 was only a proposal, not a final map. He said, “We needed to get something out there. And this is something Senator Seliger and I were working on and off with – during the session with different areas of the State. And when we realized we [were] going into special session, we tried to put this together so we would have something for special session.” Id. at 4- 5. Solomons laid out Plan C125 (H.B. 4). He stated, “Throughout this process I’ve appreciated the feedback that I’ve received from the members of the body, senators, congressmen, members of the public. And I would regret to tell you that not all congressmen provided me with proposed maps for their districts, including the Honorable Lloyd Doggett.” Id. at 8. (A) Solomons laid out the locations of the four new districts and said, “Two of the four new districts, Districts 34 and 35, are Hispanic majority districts, with the Hispanic voting age population of each district exceeding 50 percent.” D-601 at 9. Solomons said they had not created a Hispanic majority district in North Texas because (1) none of the 162a proposed maps had over 50% HCVAP, whereas they were able to draw a majority HCVAP district in Central Texas (CD35), (2) they “drew a coalition district in District 6, with a combined Hispanic voting age population 39.0 percent and a black voting age population 11.4 percent, which is exceeding 50 percent”; (3) “we were concerned that the proposed maps did not maintain communities of interest. For example . . . Representative Veasey’s map [Plan C121] split the cities of Fort Worth, Haltom City, Arlington, Grand Prairie, Dallas, Irving, Farmers Branch, Carrollton, and Boggs Springs. Most large cities get split, but we tried to avoid splitting smaller cities whenever possible. Our North Texas districts keep Arlington, Grand Prairie, Haltom City, Farmers Branch, Carrollton, and Boggs Springs all whole. We actually tried to keep small cities and small counties whole throughout the State when we were able to.” Id. at 10. Solomons stated, “It’s a work in progress. We have to put something out there and that’s what we put out here with HB4. Mr. Seliger and I both have the same maps to put out as a proposal. My goal is to pass – and I think our goal is to pass and Legislature’s goal is to pass even with critics involved and everybody that has their own opinion is trying to pass a fair and what we think is a legal map which represents the people of Texas. And I hope to have the opportunity to listen to the members of the committee and the public 163a so we can find ways to improve the map.” Id. at 11. (B) Regarding the time-frame, Solomons stated, “We are under a timeframe in the special session. We don’t have months to do this. But we do – this has been discussed – I don’t know – from before session to throughout session to now with different ideas and different thoughts on this entire process. So now we’re reaching a point where we’re kind of filtering down into some sort of proposal that we – that we believe and we would like to try to think that we’re drawing a fair legal map.” D-601 at 11. Solomons continued, “I have not made any legal determination as to whether a minority majority district is performing and, therefore, protected under the Voting Rights Act. That will be part of the process going forward, the process which I had hoped that all members decide to participate in. With that said, to determine straight minority majority districts I look at total population percentages of blacks and Hispanics. We double-checked that these districts were still minority majority districts in the voting age population, the SSVRs and Hispanic citizen voting age populations. There were seven minority majority districts with a Hispanic citizen voting age population above 50 percent, but not all of them were districts that were able to [consistently elect a representative of choice]. The proposed map, HB 4 includes eight Hispanic majority districts. The map also maintains three black opportunity 164a districts with one [coalition] district . . . being District 6. The overall population growth in Texas has been 20.6 percent, and it’s assimilated throughout the state, mostly to the urban areas.” Id. at 12-13. Solomons noted that the Hispanic population had grown 41.8%. Id. at 14. He stated that they had already had comments about CD36 in Harris County. Id. at 15. (C) Rep. Alvarado (Hispanic, Democrat) submitted a statement on behalf of Congress- woman Sheila Jackson Lee expressing her disappointment that the proposed map “does not reflect the major growth that occurred amongst African-Americans, His- panics, and Asians in the last decade pursuant to the 2010 census in the State of Texas” and that “the plan is retrogressive and appears to violate the Voting Rights Act.” D-601 at 21. Alvarado stated that Congressmen Al Green and Gene Green were also opposed to the map. Id. Solomons noted, “[L]et me make sure everybody understands the timeline on this. We’re in special session. Anything that they want to get to the committee needs to be done fairly quickly. We don’t have weeks and two weeks and five weeks and months to kind of go through this process.” Id. at 22. (D) Rep. Menendez (Hispanic, Democrat) spoke on behalf of himself and Congressman Charlie Gonzalez, stating that a minority opportunity district should not be based strictly on 50% numbers, but on whether it is effective. D-601 at 23-24. He continued, 165a “So given that premise, I would like to speak briefly about the 20th Congressional District. Foremost, it was the first district in the State of Texas to elect an Hispanic to Congress in 1961. It has always been the seat that remains entirely within Bexar County, and it also represents the heart of the people of San Antonio physically and culturally. And we would like and feel that we deserve to have a district that solely represents their interest in our community. The proposed [HB 4] does not do that. Furthermore, it has also been largely an urban district that has been compact. And, finally, it has always been a majority Latino district that has effectively elected the candidate of their choice.” Id. at 24. He further stated, “The 2010 census also indicates the 20th Congressional District needs to lose about 13,217 people in order to achieve the ideal population of 698,488. Because of this our community asserts that additional congressional districts can be created around the 20th Congressional District with minimal impact to the current lines. The proposed HB 4 map changes the district in very significant ways. First of all, it decreases the percentage of registered Hispanic voters below 50 percent. It takes out for the first time in its history, it takes out the cultural soul of the district by removing large parts of Edgewood. Edge- wood being the school district that started Edgewood versus the State of Texas, and the West Side, which are core communities 166a of interest of the current 20th Congres- sional District, which I’ve also represented in this State House for the last ten years. And it puts the district and it stretches that district all the way to South Austin. Chairman Solomons and members, our family business was built by the family and has always been represented by Gonzalez. It would also be placed in this new district that comes up to South Austin. My father when he chose his final burial place, he chose close to his favorite restaurant in this community now would be in an area represented by someone that possibly could come from South Austin. I respectfully ask you and this committee to maintain the integrity of the 20th Congressional District as much as possible when creating additional districts and making changes to the existing districts in Bexar County.” Id. at 25-26. When asked by Rep. Hilderbran what percentage was necessary for an effective district, Rep. Menendez said, “In order for effectiveness on election day I personally believe it needs to be in the 56 percent, not 51, not 52. But I think it needs to be 55 plus.” Id. at 28. In response to Rep. Menendez’s concerns, Solomons noted that “District 20 remains a Hispanic majority district in Bexar County. Its Hispanic – Hispanic citizen voting age population was 63.8 percent. . . . And the new map dropped it to 57.0 percent. But that change was actually necessary to create a new Hispanic majority district, District 35. And our – our folks who looked at all that, but from the 167a legality side and retrogression and so forth, don’t – do not believe it’s retrogressive. And a retrogression analysis shows that the Hispanic community still elects a candidate of its choice in nine out of ten elections. So even though it did drop, we created a new Hispanic majority district and it’s – and they could still elect – still elects a candidate of its choice in a retrogression analysis.” Id. at 32. Rep. Menendez said, “I definitely appreciate the fact that you have looked at the issue of retrogression. I’m glad that the committee is looking at it seriously, and I look forward to working with y’all.” Id. at 33. (E) Rep. Alonzo (Hispanic, Democrat) stated that he was “extremely disappointed” with the lack of a new Hispanic opportunity district in DFW in the proposed plan. D- 602 at 35. Rep. Veasey (African American, Democrat) noted that his proposed map had a DFW district with 66% HVAP. Id. at 38. (F) Rep. Naishtat (Anglo, Democrat) of Austin spoke and asked that as much of Travis County be placed in one district as possible. D-601 at 39. He noted that it did not serve Austin or the rural populations that Austin is joined with to have them in districts together. Id. at 39-40. He further noted that Austin and Travis County regularly elect minorities and “[i]t does not serve our minority communities to be linked with representation outside of Austin; for example, with minority communities in San Antonio.” Id. at 40. 168a (G) Rep. Walle (Hispanic, Democrat) from Harris County spoke against the proposal, arguing for a new Latino opportunity district in Harris County. D-601 at 42-43. He also noted that minority voters in Harris County act in coalition. Id. at 43. Rep. Alonzo agreed with the need for a new Latino opportunity district in Harris County. Id. at 50. (H) Rep. Veasey complained that all the Afri- can Americans were put into one district in Dallas County and did not need to be. Id. at 56-57. He noted that although minorities accounted for the growth and the population for new districts, they lost strength under the proposed map, while Anglos gained strength, and that the proposed map violated the VRA. Id. at 65. Solomons stated his opinion that the map was an improvement for minority representation. (I) Cynthia Garza spoke on behalf of Congress- man Hinojosa expressing disappointment that the Rio Grande Valley did not get a new seat despite the growth there. D-601 at 72-74. Hinojosa supported MALDEF’s map that placed a new district anchored in Hidalgo County. Id. at 75. Elisa Alvarado also testified in favor of a new district anchored in Hidalgo County. Solomons said, “We thought it was important to have one district anchored in Cameron County and another in Hidalgo County, and the only way to do that was stretch both of them somewhat north.” Id. at 74. 169a (J) Anita Privett of the League of Women Voters testified in opposition to the map. D- 601 at 81-83. She noted that since they had only had the proposed map since Tuesday there had not been sufficient time to analyze it fully. Id. at 82. She stated, “The league is especially concerned about the minority voters in Districts 6, 7, and 27. Under the proposed map minorities with – comprise more than 50 percent of the population in each of these districts, yet would not have the opportunity to elect candidates of their choice based on 2008 election data. Previously District 27 provided minority voters the opportunity to elect candidates of their choice. That they would be unable to do so under the proposed map suggests clear retrogression.” Id. at 83. “Additionally, communities of interest are not given any weight under this map. For example, the proposed Dis- trict 27 appears to incorporate areas in different parts of the state, as well as portions of counties, thus fracturing and diluting the collective influence of communities of voters residing within it.” Id. Regarding public input, Privett said, “There’s been public input in general about what people want to see in their maps. But, you know, you’ve got to have a real map to be working with. And we only got your – this particular map Tuesday morning. And I’ve seen no evidence personally that any of the input that we gave – and I testified at a number of hearings since the Legislature started. I’ve not seen any evidence that my input 170a has had any impact on any map.” Id. at 86. Rep. Hilderbran responded, “I think there’s been discussions across the state that we had those hearings and they’re firmly planted in this map. It may have just not been the one you wanted or the one that the people you favor to do it. But I see that in West Texas, even this new district in Tarrant County, and other places. I see that input that we heard. So there’s maybe differing opinions, but, I mean, I think there’s been opportunities.” Id. Rep. Alvarado stated, “Thank you for making that comment because I think what the frustration has been with some folks is that when we were on the road and people were testifying, we weren’t testifying on a proposal or any type of map hoping that whatever testimony was given, that would be reflected in the map. And there are a couple of things that stick out in my mind. Houston had the largest turnout, well over 100 people. And there was a consistent message about creating an additional Latino opportunity district. That’s certainly not reflected in the map. There was a lot of testimony here in Austin about keeping as much of Austin and Travis County together. That’s not reflected in the map. Another opportunity district in the Dallas and North Texas, and that’s not reflected here. So you kind of wonder where the input from this map came.” Id. at 87.

171a (K) Joey Cardenas of LULAC testified in opposition to the map, stating that Latinos should get all four new seats and that other proposed maps were preferable. D-601 at 88. (L) Bill Burch testified in support of the plan submitted by the Grass Roots Institute of Texas (“GRIT”). D-601 at 100-02. He testified that approximately 709,000 of the population growth was HCVAP and “the increase in the Hispanic citizen voting age population from 2000 to 2010 was an increase of 22.4 percent to 24.7. So the last ten years has resulted in an increase of 2.3 percent for the Hispanic population citizens.” Id. at 102, 107. (M) Rep. Alvarado stated, “I want to make sure it wasn’t done like the last time because at the – when we did the – our House legislative districts we presented our amendments and we had amendments that affected the proposed map and then we were being rushed to vote on things at the same time.” D-601 at 111. Alvarado continued, “I would just like to ask on the record if we’re going to take more testimony like this and allow members to submit amendments, and then there would be another hearing where we would – or another meeting where we would actually vote and have time to digest what those amendments were?” Id. at 112. Rep. Alonzo agreed, “I would like to add on to what Representative Alvarado just said. And, also, just because I have concern with just about the way the whole process has kind of taken place 172a and the fact that – particularly with the minority members that serve on this committee that – I don’t think that our input has been taken enough in regards to this – to the plan that was laid out today. . . . And I would like to request to the chair that – first of all, that we get a lot more notice the next time we get ready to have a hearing because I know that a lot of folks from Fort Worth would have liked to come down to testify, but they did not have enough time to do that today and that – also that the next hearing that we have that we also have it in a committee room like we’re having it today and so people can come down and actually testify.” Id. at 112-13. (N) Ray Guerra of the Greater Houston Civic Coalition said he found out about the hearing “at the last minute” but was able to come, and stated that, given the tremendous Hispanic growth in Harris County, it should be given another Latino opportunity district. D-601 at 115-16. He further said that he “didn’t get a chance to glance at the map, it’s been just over 24 hours, I don’t think we’ve – the community has had an adequate opportunity to analyze the map and I don’t think we’ve had an opportunity to sort of organize people to come and provide input. And so we do just want to agree with the representatives that have said so that we do look forward to receiving more notice . . . .” Id. at 116. (O) Rep. Alvarado asked what metrics were being used in drawing the proposed map. 173a Solomons responded, “[P]robably the top four numbers you look at are Hispanic total population, Hispanic voting age population, your Spanish surname voting registration, which is SSVRs, and your Hispanic citizen voting age population. Those are the four. And, in fact, the last one is probably one of the more important ones, if not the most important. But it’s certainly one of the more important ones. But those were the four things we were probably looking at in trying to look at these districts.” D-601 at 121. Solomons continued, “They’re based on the US Census data and those are the numbers you use. And we didn’t get, actually, the Hispanic citizen voting age population numbers until just not too long ago. . . . But those are the numbers that – based on Voting Rights Act considerations, those are primary numbers that you look at, especially the Hispanic citizen voting age population. And that’s what my general counsel told me, so that’s what I’m passing along on to you-all.” Id. at 123-24. “And I will tell you that general counsel tells me and [Leg] counsel tells me the Hispanic citizen voting age population is important, one of the most important of those four items. That’s all I can tell you. You know, it’s not like I’ve got 20 years of history with redistricting. I was asked to run – I was asked to do this. All I wanted to do was be on the committee.” Id. at 125. Rep. Alonzo said the map was retrogressive and violated the VRA, 174a and Solomons disagreed, stating that counsel had said it met the requirements. Id. at 126. (P) Rep. Veasey said, “The district [CD34 in DFW] that I drew on [Plan C121] is – 100 percent follows the Voting Rights Act. I want to be as crystal clear on that as possible. It’s an effective Latino opportunity district that is 71.8 percent Hispanic with the Hispanic voting age population of 66.2 percent. And if – and when you add that with a minority citizen voting age population, it puts it at 65 percent. And I just think that that’s important to point out because I’m not sure exactly what [Downton] is looking at . . . “ Id. at 128. He argued for his plan as being consistent with the Republicans’ argument in 2003 that they should have the same percentage of seats as their percentage of the statewide vote. Id. at 169-71. (Q) When Rep. Alonzo asked Solomons if he would make the general counsel (Downton) available to HRC members, Solomons responded, “He’s not your general counsel. He’s my general counsel. And he helps – and he takes input because I’ve asked him to take input from any concerns people have. But he’s my general counsel.” D-601 at 131. Solomons did not allow HRC members to ask Downton questions during the hearing. Tr1609 (Solomons) (“I felt like he was my general counsel, first and foremost, and the general counsel for individuals, 175a confidentiality with members; in particular, you know, if they talked to him, as committee members or members, that he had confidentiality issues that he had to abide by with those members . . . . I didn’t think it was appropriate for members of the committee to put Ryan Downton, as a map drawer and counsel for the committee, to be asking him questions in that hearing, yes, at the time.”). (R) Rep. Alonzo submitted “for the record a list of 17 people that had requested they create a North Texas opportunity district.” He also asked whether they would get 48-hour notice of a vote on a map, and whether any amendments or proposed substitutes would be submitted at least 24 hours before that meeting, and Solomons answered, “I believe that’s correct.” D-601 at 133. (S) Numerous witnesses testified in opposition to the map and its treatment of Travis County. D-601 at 134-36, 148-49, 150-54, 157-58, 159-60, 160-62, 165-68, 189-91, 198-200, 217-19, 219-21, 221-23. At the time, Travis County went into six districts, with only four residents in CD31. D-601 at 136. Solomons said they were trying to fix the four residents in CD31. Id. at 138. Witness Deece Eckstein emphasized that Travis County voters did not come close to being 50% of any district and that Travis is the largest county in Texas and may well be the largest county in the country that does not have the opportunity to elect its own member of Congress. Id. at 141. 176a (T) Rep. Alonzo asked if Interiano was available, and Solomons stated, “His assignment is for redistricting.” Alonzo asked, “And since he is part of the house is he available for us to work – for him to work with us regarding the redistricting?” Solomons answered, “Certainly. Give him, me, Ryan, anybody your comments to what you would like to see or not see. We’re all available to be accessible for your comments.” Alonzo asked, “Okay. And do we have somebody on the committee staff that’s available as an attorney to assist us in redistricting – in the redistricting committee?” Solomons said, “As an attorney to – [Downton] is an attorney and he is the general counsel for me and is here to help me as chairman with everybody.” Alonzo asked how Downton was paid, and Solomons said, “His salary comes out of the redistricting committee’s budget.” Alonzo asked, “And that being said, so he can communicate with us regarding, you know, suggestions about proposals we want to bring before the committee?” Solomons responded, “Sure. I mean, in the context of – as you know with the other maps, anybody who had comments that came to a committee, not just through me but through [Interiano] and others, we were all taking in everyone’s comments. And he’s there to help take your comments and tell you if that – if he thought that might be something you would want to work with or not. But his – you know, he has an attorney/client relationship with me.” Id. at 146-47. 177a (U) Bill Owens testified in favor of his own map, C111. D-601 at 173-76, 186. Owens also said, “The comment was already made that we had 24 hours notice of this hearing, which makes it very hard for people from distant communities to come here. And secondly, we’ve only had 48 hours notice of the map. We really ought to do better.” Id. at 178. Janice Banks ofTexas NAACP testified against the map and noted that there had not been sufficient time to review it. Id. at 224. (V) Sen. Gallegos said that the proposed map was “full of the Christmas turkey.” D-601 at 229. He argued that the map did not create sufficient minority opportunity given the minority growth. He and Veasey discussed that when minorities are only a small portion of a district, the representative does not feel the need to work for their votes. Id. at 238. 179. Plan C130, Seliger’s statewide substitute plan to C125, was made public on June 2 at 6:45 p.m. Changes from C125 included: CD1, CD5, CD14, and CD36 in East Texas; CD21, CD25, CD10, CD27 in Central Texas; the border between CD20 and CD35 in Bexar County; CD15, CD27, CD28, and CD34 in South Texas; and districts in and around Harris County/ Houston (including CD7, CD14, CD22, CD10, CD36 (no longer a horseshoe), CD8, CD2, CD9, and CD18). Seliger stated that they were considering Plan C130 as a committee substitute to respond to phone calls and suggestions made about Plan C125. D-602 at 24. 180. Later on June 2, Opiela emailed Interiano about a concern that the Woodlands was removed from 178a CD8 and put into CD10 in Plan C130 and asked for the best way to fix this. PL-311 Pt. 5; D-696. Interiano responded to Opiela and copied Doug Davis and Downton, “I just talked to Doug [Centilli] and suggested that he work with Eric to draft an amendment and have one of the Senators offer it in Committee. I think this was an oversight on our part. Ryan let me know if I’m missing anything.” Downtown wrote, “If it was taken out, it wasn’t intentional. But follow up with Dougon one specific request in that area.” 181. Opiela then emailed Interiano with the subject “cd20” and asked, “Is the regression analysis back on C130? does it perform 10 of 10?” D-616. Interiano responded, “Don’t know if Ryan requested it from the AG. And it would take several days for the AG to do it. Either way, I’ll put in the request now.” Later, Opiela emailed Interiano asking if he could give him the political and HCVAP reports for Plan C130. PL-311; NAACP-620. Opiela said he would rather use RedAppl data than his data because “your numbers are higher, due to the way RedAppl splits vtds (it doesn’t take into account turnout when disaggregating vtd data to the block level).” He continued, “If not, that’s fine, but I’m trying to prevent the grousing about how low cds 10 and 21 are . . . .” Interiano sent Opiela the reports on June 3 at 8:10 a.m. Id. Also on June 3, Opiela forwarded Downton a proposed change from Lamar Smith to put a specific precinct into Pete Olson’s district (CD22). D-697. 182. On Friday June 3, 2011, the SRC held a public hearing on congressional redistricting in the Capitol Extension. D-602 (transcript). Seliger laid out S.B. 4 and explained the bill. D-602 at 11. (A) Seliger stated that no decisions had been made about how to handle amendments. 179a Id. at 26. He said the over-arching principles were that the map be fair and legal and comply with § 2 and § 5 of the VRA and “no consideration is given to changing a line, or changing a district, or a shape, that it, i—it isn’t vetted legally. The people who draw the lines are lawyers. This Com- mittee has retained some of the most dis- tinguished specialists in Redistricting anywhere in the country, and followed their advice.” Id. at 11-12. He further stated, “There are four new districts, and placement fell where the population growth was most significant. And, it should come as no surprise that it’s along the I-35 Cor- ridor [and] the Valley. District Number 33 is in the Dallas-Fort Worth area, is an Arlington based district and includes Par- ker County and a portion of Wise County. Section, District 34 is in the Valley. It is a Valley bia—based Latino opportunity district combined from Cameron and Hidalgo Counties. District 35, in Central Texas, is a new Latino opportunity district based in San Antonio, traveling north to Travis County. The concept from this district came to us from MALDEF’s Plan 122. And, District 36 is in Harris and South- east (sic) County.” Id. (B) When asked which Congresspersons had input into the plan, Seliger listed the following as “ones who have come to the Cap- itol” or those with whom he had spoken: Barton, Conaway, Al Green, Jackson Lee, Neugebauer, Flores, Olson, Canseco, Doggett, McCaul, and Smith. When asked 180a if he had conferred with minority members of the Senate in developing Plan C125, Seliger said he thought he had spoken to almost all of them. Id. at 15. Sen. West stated that he had not had any input into Plan C125 and in fact had not seen it before it was published. Id. at 15-16. Sen. Zaffirini stated, “Chairman Seliger, I’ve been on every Redistricting Committee since my election in 1986. And I must say that I have never had less input into the drawing of any map than during this Session.” Id. at 21. She asked if any members of the Committee or member of the Senate were actually involved in drawing the lines, and Seliger said, “No, well, . . . it is very much a product that I developed with my counterpart in the House [Solomons]” and the staff drew the map lines on the computer. Id. at 22. When asked to specify the staff, he named Doug Davis and “a gentleman in Redi- stricting in the House, who is also an attorney who provided substantial input.” Id. at 22. When asked about attorneys consulted, Seliger stated they consulted regularly with TLC and committee counsel (Professors Guinn and Morrison, and Bob Heath of Austin). Id. at 22. He also stated that they had talked with individuals from the Governor’s office, Lieutenant Governor’s office, and the Speaker’s office. Id. at 22. When asked if any members of his legal team had provided any analysis in writing, Seliger responded, “No, Ma’am, 181a I have none, I have no written stuff.” Id. at 23. (C) When asked about whether CD23 was still effective, Seliger responded that it had a total SSVR of 52.8%, which “reflects the Hispanic population in District 23.” Id. at 24. Sen. Gallegos criticized the map as violating the VRA and having only ten districts that could provide minority electoral opportunity. Id. at 28-29. Sen. West read into the record a statement from Sheila Jackson Lee expressing her opposition to C125 because it did not reflect the minority growth and violated the VRA. Id. at 30. When asked about minority involvement in the plans, Seliger noted that Interiano was Hispanic but none of the Senate staff or legal team were minority. Id. at 31. (D) Sen. Watson (Anglo, Democrat) testified in opposition to Plan C125 because of its treatment of Travis County, which was “sliced into five different congressional districts” and the City of Austin, which was divided into six different districts, despite its population being sufficient to have almost two whole districts of its own. Id. at 32. He noted that Travis County would not make up more than 35% of any district and that the treatment of Travis County was inconsistent with that of other large counties. Id. at 33. He also complained that the minority population of Travis County had been divided and that the map ignored their historically effective 182a coalition. Id. at 33. He further noted that the only district in which Travis County voters were a plurality of the district was CD25, which was 67.3% Anglo, and thus Anglo voters were being treated better. Id. at 35. Watson further explained that the map did not respect communities of interest, especially counties. Id. at 37. Watson noted that CD25 comes into east Austin and picks up Doggett’s home and a lot of the minority community of Eastern Travis County and puts them in a district that runs to the suburbs of Fort Worth. Id. at 39. Watson testified that C125 was retrogressive because it decreased the number of districts in which minorities could elect candidates of choice. Id. at 40. (E) Sen. Rodriguez complained that the map did not have two new minority districts, that it put half of El Paso County into CD23 with San Antonio, and that CD23 was drawn to protect Canseco rather than reflecting communities of interest and geography. Id. at 41-42. (F) Rep. Menendez testified on behalf of Con- gressman Gonzalez, complaining that the SSVR was decreased from 57% to 50% and downtown and large parts of Edgewood and core communities of interest in benchmark CD20 were removed from the district. Id. at 45. (G) Sergio DeLeon from Tarrant County testified that there was insufficient notice for people to make arrangements to come to the hearing. Id. at 48. He opposed the plan 183a because it did not create a Hispanic opportunity district in the DFW area and it split the Hispanic community of Tarrant County into a number of districts. Id. at 48-49. Chairman Seliger stated that notice of the hearing was posted “at 4:46 last Tuesday” [May 31]. Id. at 50. (H) Cynthia Garza testified on behalf of Con- gressman Hinojosa (CD15) in opposition to the map because it failed to create additional representation in the Rio Grande Valley. Id. at 68-69. Greg Gonzalez from San Antonio testified against the way CD20 had been split on race to combine Hispanic population in San Antonio with Austin in CD35. Id. at 71. Eliza Alvarado, the Hidalgo County Democratic Chair, testified against the map’s treatment of Hidalgo County and the Valley. Id. at 72- 73. Anita Privet of the League of Women Voters stated that the map was developed in secret, was unrepresentative and unfair, and violated the VRA. She also stated that there had been inadequate time to study the map. Id. at 75-77. (I) Nina Perales testified on behalf of MALDEF and the Texas Latino Redistricting Task Force in support of MALDEF’s Plans C122 and C123 (12-district plans) and against C130. Id. at 77. She complained that Plan C130 unnecessarily reduced Latino electoral opportunity in CD20 and CD23, cut up San Antonio neighborhoods, and stranded the Hispanic voters of Nueces County in an Anglo-controlled district. 184a She said, “Plan C130 reduces the Spanish Surname Voter Registration in District 20 to 50 percent. That is a dramatic drop of 8 percent. As a result, the votes garnered by Latino candidates are also dramatically reduced in the new version of the district. This reduction is unnecessary and makes it more difficult for Latinos to elect their preferred candidate in District 20.” Id. at 78. She continued, “Although Plan C130 does not reduce the percent, Spanish surname registered voters in Congres- sional District 23, it nonetheless, retrogresses that District by altering its geography and voter composition. Plan C130 retrogresses CD23 by removing large portions of the South Side of San Antonio, as well as Maverick, Zavala and Dimmit Counties, and adding in Frio, La Salle, Loving, Winkler, Ward, Crane, Upton, Reagan, Schleicher, and a portion of Sutton County. As a result, Latino voters in CD23 no longer have an opportunity to elect their candidate of choice. The changes in South Texas are gratuitous and suggest an intent to obstruct, or wholly prevent Latino voters from electing their preferred candidates in Districts 20, 23, and 27. On the other hand, the Redistricting plans offered by the Texas Latino Redistricting Task Force contain nine Latino-majority congressional districts in Texas. We believe nine districts are required by the federal Voting Rights Act to prevent dilution of Latino voting strength.” Id. at 78. MALDEF also proposed a tenth district 185a that would be a coalition district in Harris County. Id. at 79. Perales stated that she believed both C125 and C130 were retrogressive in violation of § 5 of the VRA and violated § 2 by failing to create the required number of opportunity districts. Id. at 79-80. Perales stated, “I can say that we had a full and fair opportunity to communicate our views to the Chairman of both the Senate and the House Redistrict- ing Committees” and although the concept of CD35 was taken from the MALDEF plans, specifically C122, C130 had a different configuration. Id. at 80. Perales further testified that between 1990 and 2000 Latinos were responsible for much of the growth but did not gain a new opportunity seat even though the State got two, and now that they were again responsible for the growth and the State was getting four new seats, Latinos were not seeing increased opportunity. Id. at 82-83. Perales said that CD34 was a swap for the loss of CD27, and “District 34 attempts to com- pensate for that, or be a one for one exchange, but it is oddly shaped because it has to bypass Nueces County, which has almost a quarter of a million people in it, and keep going to the north through less populated counties. It is irrational to divert the district that comes up out of Cameron County, around Nueces County, and keep going up to the north, when Nueces County is right there and has traditionally been included in that configuration.” Id. at 84. With regard to DFW, 186a Perales said, “I understand that there was some talk in the House hearing yesterday, that a, that a district could not be drawn that way because nobody had provided an . . . example of a district that was over 50 percent Hispanic citizen voting age population. I wanna make absolutely clear that such a district can be drawn. We proffered a map that was built out of whole VTDs, that was at 45 percent Hispanic CVAP, but it is absolutely possible and I know that any competent mapper who is advising the Senate Committee, can show that that district can be drawn. It is there. And, I would also like to point out that the Latino community that is encompassed in our proposed district is spread across more than three . . . districts. They are currently residing in the proposed, under C130, District 30, District 6, District 33, District 12, and District 26.” Id. at 85. (J) Lydia Camarillo of the Southwest Voter Registration Education Project and the Latino Task Force testified that the map did not provide sufficient Hispanic electoral opportunity. Id. at 93-94. Stewart Snider, the President of the League of Women Voters of Austin testified against C125 and C130 for denying Travis County a congressional voice. Id. at 100. Bill Betzen of Dallas testified against the map because it fragmented minority voters in Dallas. Id. at 101. Domingo Garcia of the Task Force testified against C125 and C130, focusing on the lack of a Hispanic 187a opportunity district in DFW and the fragmenting of Hispanic population there. Id. at 104-05. Joey Cardenas of LULAC stated, “We had plenty of time, plenty of time to talk about these maps, instead, this Committee, along with the House, and I testified yesterday at the House, as well, instead we waited till the last minute to present the map on Tuesday. MALDEF, the Latino Task Force, of which Southwest Voter Registration, MALDEF, and LULAC, and MABA, and other organizations I’m part of, we offered maps a long time ago, C122, and C123, are the maps that, that we support. In those maps, LULAC insisted that we have four, all four congressional districts be drawn up as majority Latino congressional districts. We even targeted the specific areas where that needed to occur, because of the growth that we saw in those areas in Harris County, and the Dallas, Tarrant County, certainly one in the Valley, and another one anchored in the I-35 Corridor. You know, we offered those plans a long time ago, we should of been having discussions, we should of been at the table with those types of discussions, that’s how committees work, not offering it at the last minute, in a Special Session.” Id. at 110. Deece Eckstein, the Intergovernmental Relations Officer for Travis County Texas, on behalf of the Commissioners Court, and the people of Travis County, testified in opposition to Plans C125 and C130. Id. at 116. He read a resolution into the record, 188a which asserted that the district lines destroyed a functioning crossover district (CD25). Id. He also complained about the way Travis County was split so that it was only a small part of five districts, and that this was inconsistent with the way other counties were treated. Id. at 117. When Sen. West asked Seliger why Travis County was divided that way, Seliger replied, “It was just, simply, as we went through the map, and, and put together the right size of districts, just did it that way.” Id. at 118. He also said, “Yes, as we went in from the corners of the edges of the map and we came in to get the number of people to get 698,488.” Id. (K) Sen. West then asked if Committee counsel could be consulted about Travis County, and Morrison, Guinn, and Heath stated they thought it was constitutional. They also stated that they had just seen the plan and that they had not been involved in drawing the plan. Id. at 126. West asked if they believed that the map would pass muster under § 5. Guinn stated that he would want to study past election results before saying whether the plan complied with § 5. Morrison stated that the process had been “quite different” from what they had seen in the past. He noted that they did not see a congressional plan until the regular session ended and “[n]obody has had the opportunity to study it, the way it has been done in the past, or the way you would do it ideally.” He noted that in 2003 “we went all over Texas,” 189a looked at election results, and hired experts who did regression analyses, which would be the “ideal way to do it this time.” The Committee attorneys stated that they had not seen Plan C130 until 8:00 a.m. that morning when they arrived and were given C125 that Tuesday, and they said they were provided with a limited number of election results. Id. at 140-41. They said, “We’ve worked with Doug Davis, and have worked with the Chair on the general background legal issues that are involved and, in doing redistricting.” Id. at 141. They also said they had not provided any opinion on whether C125 complied with § 5. Id. at 140. Seliger wanted to resume public testimony, stating that the attorneys were counsel for the Committee and would be available to consult with members any time. Id. at 139. (L) Lorraine Denardis testified against the treatment of Travis County and Hays County and combining a district between Austin and San Antonio. Id. at 142-43. Rey Guerra of the Greater Houston Civic Organization testified against the plan, specifically advocating for a Latino opportunity district in Harris County based on the population growth. He also complained that a new map had been released the night before and there was insufficient time to analyze the maps and give input. Id. at 144. Witnesses overwhelmingly testified in opposition to the proposed plans. Id. at 146. Mary Coppinger testified against both plans and supported drawing 190a compact districts. Id. at 164-165. She also said, “On a housekeeping issue, this meeting called with only 48-hours notice made it very difficult for, to get clarity about what the issues were, what the maps were, and there’re no handouts, it’s difficult to really examine it. I’m fortunate, because I live locally, but I know, I’m sure people in the Valley or in North Texas, East, West Texas, have a hard time getting here on such short notice.” Id. at 165. (M) Sen. Gallegos presented a demonstration map of Harris County, C127, and a similar map, C129, that created four minority districts in Harris County. Id. at 175-76. He also presented a statewide substitute plan (likely Plan C131) and moved for adoption, which was not acceptable to the author. Id. at 178. It did not pass (8 to 6). Id. at 179. (N) Seliger then laid out Plan C130 and explained that it got rid of the horseshoe district (CD36) from C125. He said, “Jef- ferson County and Orange Counties are now made whole. Congressional District 2 moves entirely into Harris County and is more compact. New Congressional Dis- trict 36 is a Southeast Texas district without the loop over the top of Western Harris County. . . . The counties that were split in, in CD10, between Travis and Harris Counties, are all made whole. And, Travis County, where it was in six districts in C125 is now in 5. . . . From the House com- 191a mittee testimony yesterday from Repre- sentative Menendez, we have raised some of the matrix percentages for CD 20, matrix percentages are things like His- panic citizen voting age population, that sort of thing. More of the City of Mission is put into Congressional District 15 to offset changes in Bexar County with Congressional District 28.” Id. at 180. Sen. Williams proposed Committee Amend- ment 1, a technical amendment to swap a few areas between CD8 and CD10, and it was adopted. Id. at 182. Sen. Uresti proposed amendment 2 to the DFW area, which did not pass. Id. at 183-84. Commit- tee Amendment 1 was rolled into C130. The SRC voted Plan C130 (Plan C136 with amendments) out of committee on June 3, 2011 by a vote of 9 to 6. D-22 at A-1; TrA260; PL-221. A racially polarized voting analysis from the OAG showed that in Plan C136, Hispanic performance in CD20 was 9/10, compared to 10/10 for Plan C100. D-141. 183. On June 4, Denise Davis sent an email to Interiano with the subject “Re: CD20 Gonzales.” PL- 1623; US-182 at 6; US-621; US-624. Davis asked, “Can you tell me how the tweak for that’s going?” Interiano responded, “Not sure that there is going to be a tweak. Eric had told me that the RNC was working on something but that they couldn’t get it to work. We’ve been talking to the AG daily and have asked them to review CD20 again but they don’t appear to be concerned. I know that Eric is. Hope that helps but feel free to call me if you want to discuss further.” Davis replied, “Lamar called Speaker and was worried. What 192a do Phillips and Cooper think?” Interiano answered, “Last I talked to Phillips [an attorney at Baker Botts] he was on with it. The reason why the numbers drop in CD20 is because a new district is created. But it is still a performing district which the AG has reviewed. But why don’t I see if he can come in on Monday and we can discuss it in more detail.” Davis then wrote, “Ok.” 184. On June 4, Opiela emailed Interiano asking if he knew when the Plan C130/136 regression analysis would be ready. D-617; PL-311 Pt. 5. Interiano responded, “No clue.” 185. On June 6, Opiela sent Downton an email and proposed map, stating “Here’s a change of bexar county dists which will get CD20 to 56 ssvr while keeping CD35 at 44 ssvr.” PL-311 Pt.8 at 445; D-618. It had an attachment C136-CD20Mav.zip. D-618. He also sent an email titled C130-Houston Fix with an attachment called “Plan 130 Revision Suggestion” and wrote, “What I talked to you on the phone about Friday.” PL-311 Pt.8 at 446. 186. On June 6, Interiano emailed Tom Phillips and Radney Wood to see if they could meet to discuss the congressional map. PL-1670. Interiano then asked Downton and Hanna if they could meet with them “to go over the issues that are continuing to be raised re CDs 20, 23, and 29?” Id. Hanna asked “what is the problem w. 29?” Id. 187. On June 6, 2011, there was a debate on the floor of the Senate (on second reading) to consider Plan C136. D-22 (Senate Journal); US-753 (Senate Jour- nal); D-605.1 (Senate Journal). (A) Seliger laid out Plan C136, stating that it placed the four new districts where the 193a population growth was the most significant, including the DFW area: (1) CD33, an Arlington-based district that includes all of Parker County and a portion of Wise County; (2) CD34 in the Rio Grande Valley, a Latino opportunity district combined from Cameron and Hidalgo Coun- ties; (3) CD35 in Central Texas, a new Latino opportunity district based in San Antonio traveling north to Travis County (the concept came from MALDEF’s Plan C122); and (4) CD36 in Harris County/ Southeast Texas. Seliger moved passage to engrossment. Seliger said that the benchmark had seven Latino opportunity districts and three African-American districts, and that Plan C136 had eight Latino and three African-American districts. (B) Sen. West (African American, Democrat) asked if it would surprise Seliger “that most ethnic minority Democrats will say they didn’t have input nor did they see this map until you laid it out?” and Seliger responded, “No, it would not.” D-22 at A-3. West asked why no minority seat was created in the DFW area. Seliger stated that they tried to create a minority district in DFW and “carefully analyzed” the MALDEF suggestion, but they could not produce a performing district there and could not produce a district that met the requirements. Id. at A-5. He stated that the shape of the district in MALDEF’s map was “odd” and “would not be required” by the VRA. Id. Seliger acknowledged that the map did contain a number of districts 194a that could be considered oddly shaped, but he thought they were not as oddly shaped because they lacked the “tentacles reaching out.” Id. at A-6. Seliger also responded to questioning about whether CD30, the African-American district in Dallas County, was packed, and denied that it was. Id. at A-8. Sen. Hinojosa stated that the “new” CD34 in the Valley was not really new, but was really the old CD27. Id. Sen. Rodriguez stated that he was never consulted on the congressional plan, and Seliger responded that his office was open and Rodriguez did not ask to be consulted. Id. at A-9. He stated, “Well, I was like everybody else, frankly, under the impression that somebody’s working on these things, but nobody seems to have any information as to when, where, how, or anything until the plans are actually revealed. So, that’s at least my experience with it.” Id. Rodriguez complained that CD16 and CD23 split communities in El Paso and there was a potential violation of the VRA. Id. at A-10. Sen. Zaffirini asked if any minority senate members had been consulted in developing C125, C130, or C136, and Seliger stated, “Not that I recall.” Id. at A-12. She stated that she had been on every Redistricting Com- mittee since 1987 and the practice had been that the Chair invited and brought together the members to participate, not wait to be approached. Id. at A-12-13. She stated that she, as a member of the committee, did not have any input into the 195a congressional plan. When asked if any minority organizations had participated in developing the plan, Seliger stated that they had taken MALDEF’s suggestion to create “the district that goes between Bexar County and Travis County,” but that the NAACP was not involved. Id. at A-14-15. When Zaffirini noted that Nina Perales of MALDEF had testified against the plan, he stated that there would be thorough judicial scrutiny and “It’s going to the courts anyway.” Id. at A-15. Zaffirini also noted her belief that Travis County had been targeted, both in the Senate map and the U.S. House map by being split among districts. Id. at A-16. She asked if any Hispanics on the Senate side were involved in developing the plan, and he said no. Id. at A-17. Zaffirini and Seliger also discussed whether the three attorneys hired by the Committee had provided advice that the proposed plans complied with the VRA, and Seliger stated that they had not, but that the plans had been looked at by other attorneys (including Doug Davis, Downton, Interiano, and the OAG) and he was satisfied that it was legal. Id. at A-11-12; US-723. (C) Sen. West proposed Floor Amendment No. 2, which was Plan C121 (the “Fair Texas Plan”), a statewide substitute. He stated that the map preserved all ten existing minority opportunity districts and created three more, to have nine Latino opportunity districts and four African-American opportunity districts. D-22 at A-18. He 196a noted that all of the proposed Latino districts except one were majority-HCVAP and all 13 districts had a majority-minority CVAP. Id. at A-20. Seliger moved to table, stating that the map “essentially redraws the map of the State of Texas.” Id. West stated that the “congressional map has been the most closed process that I’ve ever been involved in. No ethnic minority had an opportunity to have input into this particular map. The fact of the matter is, when we had a hearing, even the attorneys that we hired to assist us did not have an opportunity to review this before it was laid out. At least, that was their testimony.” Id. at A-20-21. The motion to table passed with all minority members voting against. Id. at A-21. (D) Sen. Gallegos proposed Floor Amendment No. 3, which was statewide substitute Plan C131. He stated that the committee map violated the VRA and failed to recognize minority population growth because most districts would be effectively controlled by Anglos. He asserted that C131 better provided minority representation and was “more attuned to issues associated with communities of interest.” D-22 at A-22. Sen. Lucio supported Plan C131, stating that it showed that three of the four new districts could be minority opportunity districts and three districts could be anchored in the Valley. Id. at A-23. Seliger moved to table. The motion passed, with all minority members voting against. Id. at A-25. 197a (E) Specific changes (Amendment No. 1/Plan C137 and Amendment No. 4/Plan C140) requested by Anglo Senators on behalf of districts with Anglo Congresspersons were adopted. Seliger moved passage to engrossment, and the motion prevailed 18 to 12, with all minority members voting against. D-22 at A-26. 188. The Senate reconvened later on Monday, June 6, for third reading of C.S.S.B. 4. (A) Sen. Lucio thanked Seliger for being open and accessible to him on redistricting. D- 22 at A-27. He also stated that he was very disappointed with the map, though he was grateful that CD34 is firmly anchored in Cameron County so that the Valley has a better opportunity to elect a representative from that area. Id. at A-28. However, he felt that the rest of the map was unfair and in violation of the VRA. Id. (B) Sen. Watson complained about the map’s treatment of Travis County, splitting it into five districts and “ignoring completely and totally the concept of compactness and communities of interest.” D-22 at A-28. He complained that Travis County did not make up more than 24% of the population of any district, which was different from how other large counties are treated, and that the map discriminated against minority community in Travis County that has worked to develop a coalition that elects candidates of choice. Id. at A-28-29. Sen. Uresti and Sen Zaffirini again complained 198a that the map was retrogressive. Id. at A-30. (C) Seliger moved final passage, and it passed 18 to 12, with all minority members voting against. D-22 at A-31. Plan C141 is the Senate engrossment. 189. On June 6, Bruce sent an email stating, “Moments ago the Senate passed on second and third readings, SB4, the Senate Companion for HB4 which we heard in a public hearing last week on Congres- sional districts. We expect the bill to be received by the House tomorrow morning and be referred to our committee first thing during session. Chairman Solomons will accept possession of the bill on the floor and will make an announcement of a formal meeting of the House Committee on Redistricting for Thurs- day, June 9, 2011 for 9:00 a.m. in the Agricultural Museum to consider SB4. Chairman Solomons will also make a motion to meet while the House is in session just in case our formal meeting extends into the time House convenes. Under the House Committee on Redistricting’s requirements for committee amendments, we would request that committee amendments be delivered to our committee or legislative offices by Wednesday, June 8, 2011 by 9:00 a.m. Our offices will open at 8:00 a.m. that morning and we hope to turn around all amendments and reports received to your offices by 11:00 a.m. . . . this provides you more than 62 hours notice of the hearing, and with the posting tomorrow, there will be almost 48 hours notice to the public - twice as much as is required under the House Rules.” D-712. 190. On June 6, Hanna sent an email to Denise Davis about the plan to have the HRC meet in the Agriculture Museum, stating that the meeting should 199a be recorded by House AV. D-713; PL-1663. He wrote, “Last time on the House redistricting bill we were the only ones not recording and I am concerned we have no way to rebut someone else’s selective editing of what happened at the meeting. This has nothing to do with the house rules or public access but everything to do with the record used in court and in front of DOJ.” Hanna explained that the Agricultural Museum is popular among members for meetings because they can have a meeting while the House is in session and just walk down the stairs, and there is nothing wrong with having a formal meeting there. TrA1559-60. However, because he had noticed during the prior meeting in the Agricultural Museum on the House plan that some people were recording, but the state was not, he wanted to ensure the integrity of the record. Id. In response to Hanna’s concerns, Bruce sent another email stating that the meeting would not be in the Agricultural Museum, but in the Appro- priations Room. D-712. She wrote, “Please listen for Chairman Solomons’ announcement and for our official posting later today.” D-712. 191. On June 7, the Congressional bill was referred to the HRC. D-455. A notice of a formal meeting before the HRC was announced from the House floor for Thursday, June 9. 192. On June 7, Opiela sent an email to Downton and Interiano titled “20 at benchmark” with the attachment “Odessa in 23.zip.” D-620; PL-311 Pt.8 at 448. The email stated, “run this and see if it is 10 out of 10.” Interiano testified that Opiela wanted them to run the OAG report, and that he would share the results of the OAG reports with Opiela and others who asked for them. TrA356. 200a 193. On the morning of June 8, Solomons released his statewide substitute Plan C144. D-506; PL-1122. 194. On June 8, Opiela emailed Downton regarding an agreement reached between Farenthold and McCaul regarding CD10 and CD27 (that Farenthold would take Wharton County in exchange for Colorado County). PL-311 Pt.8 at 449. This change would later be offered as part of an amendment (Plan C146) by Rep. Hunter. A little while later, Opiela emailed Downton and Interiano on behalf of Lamar Smith requesting that a specific precinct (Bexar County precinct 4008) be moved from CD35 to his district CD21. PL-311 Pt.8 at 452; D-621. Opiela explained “it’s in CD35 in Plan C136. It’s the SA Country Club adjacent to SJS’s house. Giving it to lamar should help improve ssvr a fraction in CD35. You will also have to give pct 4101 to CD21 as well because it would be orphaned (357 people).” D-621. Downton responded, “I’ll work on it” and later “Change has been made.” PL-311 Pt.8 at 452; D-621. Also on June 8, Matt Leffingwell emailed Opiela with the subject “Changes to CD12.” D-622. Leffingwell wrote, “Our office that is located on 7th and Jones street in downtown Fort Worth was drawn out. There is no population there. We told Charlie Geren about this problem and he said he would work on this today but this has to be corrected.” D-622. Opiela forwarded the email to Downton and Interiano and wrote, “Heads up. Charlie Geren is going to offer an amendment to Plan C144 to put downtown Fort Worth back in the Fort Worth Congressional District rather than the Denton County Congressional district. Easy fix, you just have to keep the transit East of IH 35W and go under I 30.” D-622; PL-311 Pt. 5. Later, Opiela emailed Interiano and Downton to inform them that Congressman Marchant would oppose Rep. Madden’s proposed amendment 201a C147. PL-311 Pt.8 at 454. Plan C147 proposed changes to CD24 and CD26 in Denton County, and to CD3, CD4, and CD32 in Collin County north of DFW. Opiela later wrote, “Just heard from Cong. Sessions’s staff that he spoke with Rep. Madden and is now fine with C147. Cong. Johnson is as well. Cong. Marchant remains opposed.” Id. Downton replied, “Is there an amendment to the amendment that we can do to get Marchant on board?” Id. Downton agreed to an amendment that would not affect Collin County/Marchant’s district. PL-311 Pt.8 at 455 195. That same day, Downton emailed Interiano an OAG RPVA for districts 20, 23, and 35 in a proposed plan (XOAGC118) and for CD23 in a proposed El Paso floor amendment. PL-1665; US-626; Quesada-241. The RPVA summary shows CD20 dropping from 10/ 10 to 9/10 in the proposed plan XOAGC118. It also shows CD23 at 3/10 in the benchmark and 1/10 in XOAGC118 and shows slightly decreased Republican performance in CD23 in the El Paso floor amendment compared to proposed plan XOAGC118. 196. On June 9, the HRC held a formal meeting on S.B. 4. PL-222; D-116 at 122; D-455. There was no public testimony at this meeting. PL-222. Solomons laid out S.B. 4. (A) Rep. Veasey and Rep. Alonzo offered the Fair Texas Plan, Plan C121, a complete committee substitute, but it failed (Ayes: Villarreal, Alonzo, Alvarado, Pickett, Veasey). Rep. Alvarado offered Plan C126, a complete committee substitute, but it failed (Ayes: Villarreal, Alonzo, Alvarado, Pickett, Veasey). Rep. Alvarado requested to have a letter retained as part of the committee record. Rep. Alonzo and Veasey 202a offered Plan C142, a complete committee substitute, but it failed. (Ayes: Villarreal, Alonzo, Alvarado, Pickett, Veasey). Rep. Alonzo requested to have his notes made part of the committee record. (B) Solomons offered Plan C144, his complete committee substitute, which was made public the day before. It was adopted. (Nays: Villarreal, Alonzo, Alvarado, Pickett, Veasey). This plan exchanged some population between CD18 and CD29 in Houston; made some changes to CD15, 28, and 34 in Hidalgo County; made numerous changes to CD20, CD23, CD21, CD28, and CD35 in Bexar County; changed CD15, CD21, CD25, and CD35 through Hays and Guadalupe Counties and north up into Travis County; changed CD6 and CD25 in Johnson County; and made significant changes to CD12, 33, and 26 in Tarrant County. (C) Rep. Hunter offered Plan C146 as an amendment to Plan C144, that made changes to CD10 and CD27, and it was adopted. It moved Colorado County and some population from Bastrop County from CD27 to CD10 and moved Wharton County from CD10 to CD27. (Nays: Villarreal, Alonzo, Alvarado, Veasey). Rep. Madden (R) offered Plan C147 (as amended by Plan C148) as an amendment to Plan C144, which made changes to CD3, CD4, and CD32 in Collin County, and they were adopted. (Nays: Villarreal, Alonzo, Alvarado, Veasey). 203a (D) Solomons directed the staff to incorporate the amendments into a committee substitute, and the committee substitute (Plan C149) was adopted by the following record vote: Ayes (11): Solomons, Aycock, Branch, Eissler, Geren, Harless, Hunter, Keffer, Madden, Peña, Phillips; Nays (5): Villarreal, Alonzo, Alvarado, Pickett, Veasey; Hilderbran absent. Solomons moved that S.B. 4, as substituted, be reported favorably to the full House, and this motion passed 11 to 5 (Nays: Villarreal, Alonzo, Alvarado, Pickett, Veasey). D-455. 197. Later in the day on June 9, Solomons sent a letter to the members of the Texas Legislature. D-673. He wrote, “This morning the House Committee on Redistricting met in a formal meeting to consider SB 4, the bill to redistrict the Texas Congressional districts. The committee adopted a committee substitute and three amendments to that substitute. The committee report will be turned in shortly, but until then I wanted you to know that the map representing the version of the bill adopted by the Committee is already available on RedAppl and District Viewer as PLANC149 with statistical reports. As we go through this process, I want to keep you apprised of the latest developments and, within the time restrictions of the deadlines of the special session, provide you with as much advanced notice of changes to the map as possible.” 198. On June 10, a calendar rule was adopted in the House. D-455; D-663 (House Journal); TrA1082 (Hunter). Rep. Hunter (Anglo, Republican) as Chair of the Calendars Committee moved to adopt the rule governing floor consideration. The calendar rule was 204a adopted by the following vote: 146 Yeas, 0 Nays, 1 Present, not voting. D-663; TrA1084 (Hunter). Pas- sage of the calendar rule meant that amendments had to be filed by 5 p.m. on Monday, June 13, 2011. The calendar rule did not prevent legislators from amend- ing a timely filed amendment. TrA1084-85 (Hunter). Hunter testified that it is not unusual to have a prefiling deadline on major bills. TrA1084. The general purpose is to set deadlines so legislators can get their work done and get the time-frame accomplished. Id. 199. On the morning of June 10, Lee Padilla (NRCC) emailed Interiano saying, “There is no way in hell they go for TX-35 snaking all the way from Austin to Maverick County.” Quesada-68 at 50. Interiano responded, “I already told that to Eric. Is he telling you guys to draw that?” Padilla replied, “No brother, it was a brainstorm of what hasn’t been done, because honestly, everything else has. I know we disagree on the 20 thing, but as I look at the #% snake to Maverick, if you guys won’t take 20 going to Maverick, why the hell would take 35 doing it? Just venting, this whole process has been frustrating but it is what it is . . . . [Opiela] speaks for us, not Dale, Tom, Dub or any of those . . . .” 200. On June 10, Opiela emailed Downton and Interiano with a forwarded email from Kayla Sulzer (NRCC Political Specialist) stating, “Does this look better for El Paso?” PL-311 Pt.8 at 456; D-623; Quesada-68 at 30. Opiela wrote, “Still waiting for block file. Will forward when I get it.” Quesada-68 at 30. Downton responded, “How are the Hispanic numbers compared to benchmark?” 201. At 6:25 p.m., Opiela emailed Lamar Smith with the subject “tell me what you think.” Quesada-68 at 37. He wrote, “I shot this to Scott Yeldell [Canseco’s 205a Chief of Staff] and to Lee Padilla to get their opinion. I was able to keep Quico [Canseco of CD23] substantially the same GOP and get Charlie [Gonzalez of CD20] to 57.7 SSVR. I replaced the entire south side of SA inside loop 410 with Maverick and Zavala. I can get it the extra 1 SSVR point if I took W Odessa, but I know we’re not going there with Conaway. Wanted to at least get it out there as an option. I think it will probably increase Quico’s and Charlie’s Hispanic Per- formance one election. I think we should at least discuss it. If we agree not to do it, then that’s fine, but I think we should talk about it.” Quesada-68 at 37. Opiela forwarded this to Interiano on June 11. 202. At 7:12 p.m. Opiela sent an email to Downton and Interiano with the subject “C149-El Paso Amend Final” and the attachment “C149-ElPaso.zip.” D-624; Quesada-68 at 10. It stated, “Here is the block file for the El Paso amendment 51.73 McCain 52.99 SSVR compared to C149 which was 51.94 McCain and 52.57 SSVR.” TrA357 (Interiano); D-634 (shapefile). 203. At 9:59 p.m., Opiela sent Interiano and Downton an email entitled “Couple of options” that attached “two plans that should (1) improve CD 23’s hispanic performance while maintaining it as a Republican district and (2) in the case of one get CD 20 to benchmark, and the other get it within 1 point of benchmark on SSVR.” PL-311 Pt.8 at 459; D-625; Quesada-68 at 6. The two attachments are C149-CD23modEctor. zip and C149-23mod.zip. D-635; D-636. Opiela asked Interiano to “have them run against election performance.” Interiano responded, “does this incorporate the El Paso amendment you sent earlier,” to which Opiela answered “yes.” D-625; Quesada-68 at 6. Interiano separately responded at 10:10 p.m., “Only the AG and the Legislative Council can run those reports. We’ll 206a have them run on Monday but there are several other issues we are trying to address.” PL-311 Pt.8 at 459; D-625; Quesada-68 at 8. The next morning, Interiano wrote Downton, “I can’t upload this at home . . . but think this may actually be a good option since it keeps CD20 entirely within Bexar County. Curious to see what it does to CD35. Let’s discuss it tomorrow.” PL- 311 Pt.8 at 459. Downton wrote back, “Sounds good.” Id. at 460. 204. On June 11 at 7:54 a.m., Opiela sent Interiano an email with the subject “SA and cd 23 reconfigure PDF” and the attachment “CD23-C149m.pdf.” D-626; PL-1621; D-639; Quesada-68 at 38-39. The email contains the forwarded email from Opiela to Lamar Smith. At 9:34 a.m., Interiano responded, “Did you make any changes to CD35 where it would have dropped the CVAP below 50%?” D-627. Interiano testified that mapdrawers felt that CD35 needed to stay above 50% CVAP or SSVR but the Republican congressional delegation were not convinced it needed to be above 50%, and that the House and delegation had different “goals.” TrA363. Interiano also testified that the mapdrawers wanted CD20 to remain within Bexar County, and when the delegation had at one point proposed a CD20 that went outside Bexar County, they did not accept that. Id. 205. On Monday June 13, 2011 at 7:23 a.m., Opiela sent an email to Interiano (copying Downton) with the subject “optimized Friday’s Plan” and an attachment “C149-CD23modoptoep-nomav.zip.” PL-311 Pt.8 at 445; US-195 at 306; D-628; D-637 (attached map). Opiela wrote, “This is the best I could do. Only 17,360 in Maverick for CD 23. 52.4% McCain. All other districts at benchmark. Probably could get rid of all of Maverick if did Ector, but right now only districts changed are 207a 16, 23, 28, 20, 35, 15. 20 doesn’t leave bexar county, and 35 doesn’t leave Bexar in the south either. Quico has not yet seen/signed off on this, but could you please run the election performance on this one (in lieu of the non-Ector one I sent Friday). Let’s discuss this morning.” D-628. Interiano testified that this was Opiela’s proposal trying to meet the confines of the parameters “we” had set, specifically, “meeting the benchmarks and also having 20 not leaving Bexar County.” TrA364. 206. Interiano emailed Opiela and asked when he could come down to the redistricting office. PL-311; D-629; US-195 at 316, TrA313 (Interiano). Opiela responded at 8:04 a.m., “45 mins.” PL-311. As noted, the redistricting office was a shared office between the Speaker and the redistricting committee, and it was Interiano’s and Downtown’s primary office. TrA315. Interiano uploaded the plan into RedAppl around 8 a.m. The plan became strjC116 in Interiano’s RedAppl account, created at 8:15 a.m. and last modified at 8:32 a.m. TrA313-14, 366 (Interiano); US-664; D- 539.1. In CD23 in plan strjC116, counties north of the Pecos River—Ward, Crane, Upton, Reagan, Crockett, Schleicher, and Sutton—are included in CD23; Maverick County is split between CD23 and CD28; non- suspense SSVR is 54.1%; Yanez receives 50.0% of the votes in CD23 for Supreme Court Place 8; Noriega gets 46.9% of the votes for U.S. Senate in CD23; and Molina gets 45.7% of the votes for Court of Criminal Appeals presiding judge. TrA332-33 (Interiano); PL-948. 207. Interiano sent the plan to Downton. TrA322 (Interiano). Downton’s RedAppl plan list contains plan hrc1C187 (US-731NN) labeled as “Possible Floor Map” that was created at 8:33 a.m. US-658 at 3. This plan was last modified at 2:27 p.m. that day. Id. Interiano’s account also contains strjC117 (US-731HHH), created 208a at 12:45, and titled “Floor Amendment; Slight Revi- sion.” US-664. Downton’s plan hrc1C187 and Interiano’s plan strjC117 are the same except for El Paso (and slight changes in Bexar County, though the record testimony says Bexar County is the same). TrA320 (Interiano). Downton’s RedAppl plan list includes hrc1C194, titled “Solomons Floor Amendment - Final,” created at 3:16 p.m. and last modified at 3:24 p.m. US-658 at 4. This plan became public Plan C170, the Solomons West Texas Amendment. 208. At 8:16 a.m. on June 13, Interiano emailed David Hanna and asked him to go to the redistricting office. US-748. Interiano asked the TLC for and received by email a Red 106 report on the plan at 8:34 that morning. TrA314; US-754. Both Opiela and Hanna responded that they would come to the office, but no witnesses recalled the meeting. TrA317 (Interiano). At 1:15 p.m., Interiano emailed Denise Davis saying, “Let me know when I can come update you.” US-750. Davis responded at 2:04 p.m., saying, “Now.” US-750. 209. At 5:18 p.m., Interiano sent Opiela a map with the attachment “STRJC120.csv” and wrote, “per your request.” PL-311 Pt. 6; D-630. 210. Plan C170, Solomons’ West Texas Amend- ment, which affected CD11, 15, 16, 20, 21, 23, 28, 34, and 35, was made public at 4:53 p.m. 211. At 10:29 p.m. Opiela emailed Jennifer Brown (Congressman Smith’s chief of staff) and copied Lamar Smith. D-631. The email was titled “Re: okay . . . . this has to stop.” Opiela wrote, “Not that we don’t have enough to worry about, as I was making my way through all the reports for each district I finally got to the Solomons amendment. Makes me want to shoot 209a through the roof. I didn’t go through all of this for nothing today. I got the stats in CD 20 to benchmark, 58.1 SSVR, etc. just to have them drop it back down to 55.6. I had a voice wondering in the back of my head how they were able to find enough Hispanics to jack Quico up from 52.8 to 54.1. I knew they couldn’t do it alone with just 10k more in Maverick. They stole them from CD 20. This was the whole point behind this exercise. I gave them the tools to fix this, and it was used for this. I’m tempted to try to get someone to offer what I gave them as an amendment to Solomons amendment, but know that will blow things up.” Smith responded to Opiela and copied Interiano at 10:40 p.m. He wrote, “Just had a long talk w harvey hildebran who called me back. He did not know about the solomon amend and when I explained it took cd20 back to benchmark and increased hispanics in cd23 he said he wld study it and barton map and might not offer barton. Said purpose of barton was to improve cd20. He also did not kno barton wld make a lot of del unhappy and hurt canseco and took three incumbents below 55 mccain so he said he wld study solomon amend tonite before making a decision. Think if burt called him and explained his amend hh wld not offer his.” D-236; D-631. 212. At 11:06 p.m., Opiela sent Interiano an email with the subject “why.” D-235; D-632; NAACP-616; PL-311 Pt. 6. Opiela wrote, “Why do this to me? I get the stats to benchmark in 20 and look at the report tonight and y’all dropped them to 55! Please, please say this whole exercise was not for naught. Call me crazy but I could care less about 35 being at 45 – the thing will probably perform at 40, why not take that difference and attempt to meet it if we can? I calmed Lee, Holman, et al. down today to have this done?” D- 632. Interiano responded the next morning with, “It’s 210a at 56.5 . . . . and it performs 10/10. You’re looking at total when you should be looking at non-suspense VR. Call me later this morning if you want to talk further. Also, can you send me the analysis that you did on all the amendments?” PL-311 Pt. 5; D-632; NAACP-616. Opiela forwarded Interiano an email titled “Amend- ments summary” that listed proposed amendments C152-C157, C161, C167, C163-C165, C166, C168, C169, C170, C172 and summarized them. US-747. 213. At 6:59 a.m. on Tuesday June 14, Interiano wrote Smith and Opiela, stating “I’ll make sure to pass [the concerns raised in the “this has to stop” email] to the Chairman this morning. In re to CD20, it’s actually at 56.3 and it performs 10/10 so it’s a definite improvement from where we were and there is no reason to be concerned in that regard.” D-631. Opiela responded to Interiano at 8:32 a.m., stating, “No, just I think it’s not good to not have all the stats at benchmark when we know its possible. Yes, I know CD 20’s 10 of 10 performance and agree taking it below benchmark on SSVR can be justified by the creation of the new district. However, we know we can create the new district AND keep it at benchmark, so we should. Sure it performs the same, and will it make a difference in the end with the court, probably not; it just gives an easy paragraph where they can say we retrogressed, and we have to explain why it doesn’t matter. That said, today’s amendment is enormously better than C149, so let’s just play ball. I do feel I was mislead into thinking y’all were going to keep all the districts at benchmark yesterday, and frankly, I used that argument to calm Lee and Chris Homan down when they were threatening to blow it all up. What is done is done. Let’s get this bill passed!” D-632; NAACP-616. 211a 214. At 10:24 a.m., Opiela emailed Downton with “quico talking pts amend” with talking points for an amendment to Plan C170 offered by Peña (Plan C179) relating to CD11, CD21, and CD23. PL-311 Pt.8 at 463 (“returns Schleicher and the portion of Sutton that are currently in CD11 back to CD11; and moves an equivalent amount of population from Kimble County (whole) to CD21. CDs21 and 23 are balanced in Bexar County. Schleicher and Sutton have a closer community of interest with CD11, and Kimble with CD21. It also slightly improves the SSVR in CD23.”) 215. At 12:31 p.m., Opiela sent Interiano an email titled “negatively affected members.” PL-311 Pt. 5. He wrote, “McCaul (dropped to 52.35 McCain) Smith (dropped to 53.44 McCain, lost Hill Country) Canseco (dropped to 48.45 McCain, gets all of Maverick) Olson (dropped to 55.93 McCain) Conaway (loses 50% of Ector, Llano).” PL-311 Pt. 5. 216. On June 14, there was debate in the House on second reading of C.S.S.B. 4. D-23 (House Journal); Quesada-394. The current plan was Plan C149, the plan as passed out of the HRC on June 9. (A) Solomons offered a perfecting amendment (Amendment 1, his West Texas Amend- ment Plan C170, which was made public the day before) and moved passage. Solomons stated that the amendment responded to some concerns from the public and Rep. Menendez (Hispanic, Dem- ocrat) at the House and Senate hearings on the bill. He said, “Representative Menendez specifically asked us to increase the SSVR of District 20 over 55 percent. We were able to increase it to 56.3 percent. We also increased the HCVAP of District 212a 35 to 51.9 percent and the SSVR of District 23 to 54.8 percent. While we were doing this, we were able to maintain the performing nature of all the other Hispanic majority districts, and it’s acceptable to the author.” Rep. King spoke in opposition based primarily on the fact that the amendment split Maverick County/the City of Eagle Pass and also split La Salle County. D-23 at S1-S2. Rep. Villarreal stated, “this amendment is actually an improvement from what came out of committee in some very specific ways” and asked Solomons to repeat how he improved the SSVR numbers for districts 20, 35, and 23. Id. at S2. After Solomons did so, Villarreal said, “I want to thank you for making these improvements.” Id. Solomons moved passage of the perfecting amendment, and it was adopted. (B) Rep. Geren (Anglo, Republican) proposed Amendment 2 (Plan C169), which he said was “a swap between Congressional Dis- tricts 12 and 26” to make downtown Fort Worth whole, put the entire city of West- lake in the same district (CD26), and “unite the black communities in Fort Worth.” The amendment was adopted. D-23 at S3. This amendment put part of downtown Fort Worth, including Con- gresswoman Granger’s office at 7th and Jones, into CD12, and moved the African- American community of Lake Como into CD12. 213a (C) Rep. Kuempel offered Amendment 3 (Plan C172) related to Guadalupe County that would amend CD15, 27, 34, and 35 and would have kept Guadalupe County whole and moved it into CD27 (from CD15). CD27 only contained that portion of Nueces County necessary to include Farenthold’s home in CD27, and the rest of Nueces County was in CD34 going to the south along the coast, to include Hidalgo County. Kuempel stated that the constituents in Guadalupe County wanted to move it, but from his understanding, with Solomons’ perfecting amendment, it would not hold up legally. This amendment was withdrawn. (D) Rep. Alvarado (Hispanic, Democrat) offered Plan C168 (Amendment 7) for southeast Texas that would add a second Latino opportunity district in Harris County while maintaining the three existing districts. She said both districts CD29 and CD36 were completely in Harris County and had a HVAP over 50%, but the SSVR was slightly under 50%, but “both have proven to elect Latino voters’ preferred candidate.” D-23 at S7. Solomons opposed the amendment, stating that the map retrogressed CD29 because it dropped the SSVR to 35.5% and its HCVAP to only 38.6%, and the new district was only 42.5% SSVR and 41.1% HCVAP, and that it in effect had one less Latino opportunity district. Id. at S9-S10. Alvarado responded that they were coalition districts. Veasey stated that Solomons was using arbitrary numbers that “are not the gold standard” 214a and that there are other factors that need to be taken into consideration, including that the proposed district would elect the candidate of choice. Id. at S10. Solomons moved to table the amendment, which passed 94 to 47. Id. (E) Rep. Johnson (African American, Demo- crat) offered Plan C157 (Amendment 8) affecting CD5, CD30, and CD32 in Dallas County, and Plan C177 (Amendment 9) as an amendment to the amendment, which was initially adopted, but later the vote was reconsidered and the amendment withdrawn. PL-281. Amendment 8 as amended by Amendment 14 (C178) (affecting CD30 and CD32) was later adopted. (F) Rep. Veasey (African American, Demo- crat) offered his Fair Texas Plan, Plan C121 (Amendment 10). He said the plan recognized the minority growth and the “Republican doctrine” of 2003 which said that seats should be based on statewide election results. D-23 at S11. Solomons moved to table, and it was tabled. Id. at S12. (G) Rep. Alonzo (Hispanic, Democrat) laid out the Alonzo/Veasey plan (Plan C142) as Amendment 11 that created four His- panic opportunity districts. D-23 at S12- S13. Solomons opposed, stating that the new districts are not actually Hispanic- majority districts and did not reach the 50% threshold. The amendment was tabled 96 to 46. Id. at S16; PL-218 at 6. 215a (H) Reps. Turner (African American, Demo- crat) and Y. Davis (African American, Democrat) offered Amendment 12 (Plan C155), the Texas Legislative Black Caucus plan, with 14 opportunity districts (9 His- panic, 3 black, and 2 coalition). D-23 at S16. Rep. Naishtat (Anglo, Democrat) and Turner discussed the fact that minority communities in Travis County worked as a coalition, and that they did not want Travis County Hispanics carved out and placed with San Antonio. Id. at S18. Solomons opposed the map, citing “legal concerns.” Id. at S19. He asserted that it created one less Hispanic opportunity district than the committee map because CD10 and CD36 were below 50% SSVR and HCVAP. Solomons moved to table, which passed 93 to 49. Id. at S19-S20; PL-218 at 6. (I) Rep. Martinez Fischer (Hispanic, Demo- crat) offered Plan C163 (Amendment 13) to create more minority districts. D-23 at S20. Solomons opposed, stating that “it is similar in the committee map that creates eight Hispanic minority districts, but neither the new District 35 in North Texas, nor the new District 36 in Harris County are Hispanic majority districts. Neither one reaches 50 percent threshold . . . . So I don’t think the map is any better, as far as that goes, than the committee map. I don’t think that his map really does reflect the input we received from a number of folks. But I will point out that Mr. Martinez Fischer’s Plan 163 splits a 216a number of cities. All maps sort of do, but in this case, in Dallas and Tarrant County alone, it splits Garland, Farmers Branch, Carrollton, Irving, Cedar Hill, Grand Prairie, and Arlington. And that’s just a bit much for the committee and me.” Id. at S23. Solomons moved to table. Martinez Fischer said C163 creates an additional SSVR 50% district, and pointed out that the committee map cuts cities like Austin and San Antonio, so they should be consistent if that is the standard. Id. at S24. The amendment was tabled 92 to 48. Id. at S25; PL-218 at 7. (J) Martinez Fischer offered Plan C164 (Amendment 15), which he described as another statewide map that aimed to draw a new Hispanic district without affecting anybody else’s political liability. D-23 at S26. This map drew a district from Bexar County to Bastrop County that did not cut into Travis County. This map created 8 Latino opportunity districts, the same as C149, but also created 8 SSVR 50% districts, where C149 only had 7. Id. It also created 15 combined African American/ Hispanic voting age population districts, while C149 had 13. Martinez Fischer argued that the plan better acknowledged minority growth. Id. Solomons opposed, stating that, in CD36 “it still doesn’t create a 50 percent threshold. It still splits a number of cities in North Texas. And I appreciate his vision and what he thinks he wants to do in connection with arguing the legal merits of his position, but I don’t 217a think his map really takes in consideration all the work, and what we’ve done in listening to constituents, and listening to the delegation, and other members of this house. And, quite frankly, I just think it’s not the right map. And I would appreciate the members agreeing with me to move to table the amendment.” Id. Martinez Fischer stated, “The demographics and data support the fact that you can grow this map, make it more diverse. I’m bringing three examples that I can draw in my office, you know, with limited help.” Id. at S27. He argued that the committee, with all its resources, could draw a better map that increased minority representation, and they could work with his proposed map to fix their concerns about it rather than simply rejecting it. Id. He stated, “In the light of everything that seems to be screaming in this map, which is status quo, and protection for incumbency, and growing map to suit one demographic at the expense of those that are growing the map – that’s what that map does. That’s purposeful, that’s intentional, that is what we would call the way you can segregate and separate, and it does nothing to advance what 90 percent of the minorities did this last decade. It certainly doesn’t even take them into account.” Id at S27- S28. Solomons moved to table and the amendment was tabled. Id. at S28; PL-218 at 8. (K) Martinez Fischer then proposed his third plan, Plan C165 (Amendment 16), which 218a “takes a similar approach by connecting Travis County and Bexar County to create that additional district. It also maintains the two districts for the Dallas metroplex area, the one minority opportunity district in Harris County, for a total of four. Even still, on the third rendition, it performs better statistically than the Seliger/ Solomons map by taking one additional SSVR district, and it also performs better by taking the BHVAP seats and growing them from 13 in the Solomons/Seliger map and growing them to 15.” D-23 and S28. Solomons stated, “Not to belabor the point, but it’s similar arguments as the last two maps, members, and I hope you’ll stick with me and move to table.” Id. Martinez Fischer argued, “we can draw a better map that respects minority opportunity, that respects our growing diversity in this state, and it’s an improvement from what is being offered to us today. I’m not saying this is the perfect map; I’m not saying that we should do this map as is. I say we should stick to the numbers, and we should draw eight SSVR districts at 50 percent. We ought to draw 15 districts that have a combined African American/ Hispanic voting age population of over 50 percent. We ought to draw three districts where African Americans are the largest population as opposed to the one district that exists today. This better reflects and represents who we are as a state.” Id. The amendment was tabled. D-23 at S30. 219a (L) Rep. Dukes (African American, Democrat) offered Plan C166 (Amendment 17). She said it was a map that “incorporates the new Dallas-Fort Worth minority district, which is labeled CD35 in this map, that is very similar to the CD34 Veasey plan, C121, and has the same Hispanic and black performance. The Hispanic citizen voting age in this district is 45.6 percent. The black citizen voting age is equal to 18.3 percent with a combined percentage way over the 50 percent marker that Representative Solomons has been discussing. Harris County is similar to the plan presented by Representative Alvarado, Plan C126, and the new congressional district labeled CD36 has the same His- panic performance and slightly higher black performance as in Representative Alvarado’s map. The Hispanic citizen voting age is 45.1 percent, the black citizen voting age is 14.2 percent. The Austin-to- San Antonio district, CD35 in the republican plan in SB4, was moved to the Valley to give the Valley the new seat that its growth called for, but as well to keep Austin’s community of interest intact. It is labeled [CD33].” D-23 at S30. She continued, “This map keeps the minority coalition of Hispanic and African American voters in east Travis County together so their voice and vote can be effective. This is in stark contrast to SB 4’s purposeful discrimination of these minority communities. In SB4 Austin voters make up a unique community of interest with strong 220a diversity and respect for different points of view – that’s why we’re considered the heart of Texas, the oasis of Texas. In Travis County, Hispanics and African American, and Anglos act as a coalition, are able to elect the candidate of their choice from all races. Nearly every His- panic elected official in Travis County has signed a letter stating they do not want Hispanic families to be carved out of Travis County and connected with a distant population in San Antonio. . . . This plan works to ensure that African Ameri- cans, and Hispanics, and like-minded Anglo voters in eastern Travis County, eastern Austin, are able to continue a coalition and elect an individual who will properly represent their voice without the creation of purposeful discrimination as that which is done in SB 4.” Id. Dukes stated that “they have purposefully gone so far in their attempts to eliminate one person [Lloyd Doggett] that they have eliminated the voice of 688,000 people in Travis County.” D-23 at S32. Rep. Howard pointed out that Travis County is the only county of the 12 largest that doesn’t have even one congressional district in which Travis County would be more than 50% of the district population, depriving Travis County of substantive representation. Id. S34. Rep. Naishtat also spoke in favor of the Dukes plan, arguing that Austin and Travis County have elected minorities and deserve a united voice in Washington. Id. at S35. Solomons opposed, stating, “I 221a have some legal concerns about the map. It creates one less Hispanic district – Hispanic majority district – than the committee map, and neither the new District 35 in North Texas nor District 36 in Harris County are Hispanic minority districts. In fact, this map also retrogresses District 29, and quite frankly Ms. Dukes[‘] map, as I say, creates one less majority district than the committee map.” Id. at S36. He noted that in the benchmark CD29 has an SSVR of 52.6% and an HCVAP of 56%, but in C166, the SSVR drops to 34.5% and HCVAP drops to 38%. Id. Dukes responded, “I’m not quite certain where Representative Solomons pulls his numbers and sometimes I wonder about the concern put forth on minority communities, when Congressional District 29 is used as the litmus test for what is considered retrogression when one arguably can state something different.” Id. at S37. She also stated that Travis County has enough voters for a congressional district entirely within Travis County lines, and that it did not seem like the drawer of C149 “was too concerned about minority community and retrogression” because western Travis County, where the majority population is Anglo, is “entirely in one district, but eastern Travis County, where African Americans and Hispanics live, is drawn into nine separate in-and-out congressional districts.” Id. She said, “That is purposefully drawing African Americans and Hispanics into smaller portions and 222a separation to create a GOP plan of purposeful discrimination. This map is not about ensuring one member of congress does not have a voice in Washington. It is about ensuring that African Americans and Hispanics from eastern Travis County, from east Austin, will be unable to have a candidate of their choice to represent them in Washington.” Id. at S-37-S38. The amendment was tabled. Id. at S39. (M) Rep. Hilderbran (Anglo, Republican) proposed Plan C161(Amendment 18) and a perfecting amendment to Plan C161. D-23 at S39. This map increased the SSVR of CD35 to 57.4%. Id. He stated that he was trying to meet the same objectives as Solomons both politically and under the VRA, but he felt his map did better under the VRA. Id. at S40. Martinez Fischer and Villarreal opposed the map. Id. at S42-S43. Villarreal did not like how the map split up Bexar County, and Martinez Fischer felt that the split of Webb County was problematic given the Supreme Court’s decision in LULAC v. Perry that had criticized a Webb County split. Id. at S43. Solomons opposed the map, citing the Webb County issue, the split of Arlington, and the SSVR of CD20 going down from 55.6% to 54.6%. Rep. Hilderbran withdrew the amendment. Id. at S45. (N) Rep. Dukes then argued again against Solomons’ plan, asserting that it was motivated by discrimination in Austin and Travis County and that the process did not 223a allow for meaningful input. D-23 at S45- S46. Rep. Alonzo also argued against it, stating that there were § 2 and § 5 violations. He entered an exhibit asserting the minority officeholders were excluded from the process, that the Solomons map retrogressed statewide, and that there were § 2 violations for failure to draw additional minority opportunity districts as shown by Plan C121and other plans. Id. at S53-S57. Martinez Fischer stated that Solomons had “been able to sort of manipulate and be a little misleading with the SSVR and the HCVAP.” Id. at S62. Solomons responded, “I don’t think we’ve been misleading. Those are the real numbers. Now, we can argue how important they are, but they are the real numbers. They are part of a greater progression of items that you look at in redistricting.” Id. He continued, “[T]here are a number of traditional redistricting matters that you look at, not all of it is reliant on the numbers. But the numbers are very important. And that’s what we’ve relied on heavily, the numbers, as well as the other traditional items that you look at in redistricting. With the AG’s office looking behind our shoulders, and litigation counsel, and all the people that have far greater experience than maybe you or I in what you really look at to take a map, and you know it’s going to be reviewed by the DOJ or the federal courts. And so you try to draw a map that you think you can defend, that you think is fair, and that you think is legal.” Id. To put 224a some legislative intent on the record, Solomons stated, “We looked at Hispanic voting population, we looked at Hispanic voting age population, we looked at the SSVR issue, Spanish surname voting population. We looked at Hispanic citizen voting age population, the actual voters. We looked at – and for a black minority, we looked at black total population, black voting age population. But we did look at, and we had to consider, and we wanted to consider – in fact the bodies, the committees, everybody needs to look at all the various aspects, that includes communities of interest, cores of existing districts, incumbency, compactness, continuity, and other traditional redistricting principles. Whatever they are, you have a number of folks who when you look at these things you have to take into consideration, or try to take into consideration, as much of that as you can. So SSVR, although important, is not the sole factor. But it’s a very important factor. Just as is Hispanic citizen voting age population. You may have a number of Hispanics living in an area, but when you break it down to SSVRs and Hispanic voting age populations, those help make some decisions for you on where the line should go.” Id. at S63-S64. Solomons also stated they relied on retrogression analyses based on prior elections (referring to the OAG 10 RPVA). Id. at S64. (O) The House passed C.S.S.B. 4, as amended, on second reading 93 to 48. US-611; D-23 225a at S65; D-603; MALC-78 (record vote). The amended plan was Plan C182. 217. On June 15, Opiela emailed Scott Yeldell with Canseco’s office and Lamar Smith with the subject “Peña Amendment.” PL-311 Pt. 5; D-633; US-757. He wrote, “I ran the numbers based on the actual election returns, comparing the Peña amendment configuration, vs. the C182 Engrossed configuration, and found the following: [changes in Republican votes]. You would have to achieve around a 20% increase in turnout above the baseline turnout in order to overcome the decrease in performance, and that’s to break even. Lamar is hesitant to make the change if it could actually result in either no improvement or a decline in performance for Quico.” D-633; US-757. Yeldell replied “Well if the numbers don’t pan let’s pull it. Probably not going to get 2/3rd anyway.” D-633. Opiela then forwarded the emails to Interiano, Bonnie Bruce, and Lamar Smith stating, “Pull the Peña amendment.” D-633. Solomons wrote Bruce, Downton, and Interiano stating, “It needs to be withdrawn. Not going to accept the Peña amendment.” D-685; US-757. Interiano wrote to Solomons, Bruce, and Downton, “Peña said that he would do whatever we wanted him to.” D-685. 218. On June 15, the Congressional plan was considered on third reading in the House. Solomons offered a technical amendment (Plan C185) to conform the lines to existing precinct lines (with zero population changes), which was adopted, and the plan became Plan C185. D-604 at 414; D-455. (A) Rep. Zedler (Anglo, Republican) offered Plan C186 to make changes to CD6, CD24, and CD33. He said, “6 is Joe Barton’s. What it does is it takes him—he used to 226a have almost all of Arlington, the current map gives him none, or very little, and this one gives him a portion of Arlington, including his district office, as well as his home. In the current map, his home and his district office were taken out, and I think this is a good, fair way to get this situation solved.” D-604 at 415. Solomons opposed the amendment, stating, “It’s a last minute amendment. We’re not exactly sure of all its ramifications, but we can tell at this point, it could have been filed yesterday. We’d have had time to analyze it. This amendment, basically, as far as we can tell right now, splits Arlington, takes about 57,000 people out of Arlington. We made a concerted effort to try to keep Arlington whole and to keep the cities whole when possible. It does more than just the district office and an alleged house. . . . I would ask you to table this amendment, and we’ll send this bill over to the senate and see what else we need to do with the map, if necessary. But I do, dramatically, think that this is the epitome of Joe Barton, Congressman Barton, wanting to have just exactly what he wants without really going through the process, as all of the other congressmen and everybody else really did. This last- minute amendment does dramatically affect some other districts.” Id. at 418. Rep. Alonzo asked which congresspersons had input into the map, and Solomons said that both Republican and Democrat congresspersons had input. Id. at 419. 227a When asked about a new opportunity district in DFW, Solomons said there was “a difference of opinion” and “a lot of it had to do with the assimilation of Hispanics and blacks throughout the community in North Texas, and the Hispanic citizen voting age population issues. So when you start looking at those numbers, they become kind of important.” Id. Rep. Veasey asked Zedler, “[A]re you aware that in the current redistricting plan, without your amendment, that there were members’ district offices - minority members’ offices - that they lost their district offices under the plan that’s about to pass out? And you’re going to make this change, just for this one congressman, when there were minority members, African American members of the Congress, that also lost their district offices due to reapportionment. Are you aware of that?” Id. at 420. The amendment was tabled. (B) Rep. Alonzo included a petition for a Latino opportunity congressional seat in the DFW area, anchored in Dallas County. D-604 at 421-24. (C) The House passed S.B. 4, as amended (Plan C185) on third reading with a vote of 93 ayes and 47 nays. MALC-78 (record vote). Hispanic Republicans Torres, Garza, L. Gonzales, Aliseda, and Peña and African-American Republicans Carter and White voted for the map. 219. On June 15, the lawsuit Morris v. Texas, 4:11- cv-2244 (S.D. Tex.), was filed. Morris sued individually 228a as a voter in CD2 to challenge the enacted congressional plan. Morris v. State of Texas was later trans- ferred to this Court, and was consolidated with the other cases. Docket no. 72. 220. H.B. 150 (Plan H283) became law on June 17, 2011. (stipulated) That same day, the Texas Latino Redistricting Task Force v. Perry, 5:11-cv-490 (W.D. Tex.), lawsuit was filed. Plaintiffs are the Task Force and individual member voters. 221. On June 20, the full Senate considered the congressional plan S.B.4 with the House amendments. D-606. The amendments were read. Seliger moved to concur in the House amendments, and the motion passed 19 to 12. D-606 at 1194-95; US-611. Yeas: Birdwell, Carona, Deuell, Duncan, Eltife, Estes, Fraser, Harris, Hegar, Huffman, Jackson, Nelson, Nichols, Ogden, Patrick, Seliger, Shapiro, Wentworth, Williams. Nays: Davis, Ellis, Gallegos, Hinojosa, Lucio, Rodriguez, Uresti, Van de Putte, Watson, West, Whitmire, Zaffirini. D-606 at 1195. The congressional plan was reported enrolled on June 20. D-455. The plan was signed in the Senate on June 22. Id. It was signed in the House and sent to Governor Perry on June 24, 2011. Id. July 2011 222. On July 6, 2011, this three-judge court consolidated Perez, AMLC, and the Texas Latino Redistrict- ing Task Force cases. Docket no. 23. This Court also allowed LULAC to intervene. LULAC is described as the oldest and largest national Hispanic civil rights organization, with a history of promoting voting rights on behalf of Hispanics and other minorities. Gabriel Rosales and others, who are individually named along with LULAC, are members of the organization. On 229a July 13, this Court granted the Texas Democratic Party’s Motion to Intervene. Docket no. 31. 223. On July 15, 2011, Quesada v. Perry, 5:11-cv- 592 (W.D. Tex.) was filed and was later consolidated with the other cases. Docket no. 63. Plaintiffs are individual voters. The Court also granted Congress- man Henry Cuellar’s motion to intervene. Docket no. 42. Congressman Cuellar is a Latino voter from Webb County and represents CD28. He challenged the legality of Plan C185. 224. Governor Perry signed S.B.4 (Plan C185) into law on July 18, 2011. (stipulated). It would have become effective on September 28, 2011. 225. At the time the proposed plans were enacted, Texas was subject to the preclearance provisions of § 5 of the VRA. On July 19, 2011, the State of Texas filed a lawsuit seeking preclearance of the Texas House and congressional plans in the District Court for the District of Columbia. 1:11-cv-1103 (D.D.C.). The complaint asserted that the House plan “maintains or increases the number of districts at or above key thresholds for relevant metrics used to determine whether a minority voting population is able to elect its candidate of choice.” Comparing the benchmark with the enacted Texas House plan, the Complaint states that there are 30 districts with HVAP above 60% in both the benchmark and the enacted Texas House plan, 30 districts with HCVAP above 50% in both the benchmark and the enacted Texas House plans, 29 districts with SSVR above 50% in the benchmark and 30 in the enacted Texas House plan, and 11 districts with BVAP above 40% in the benchmark and 12 in the enacted Texas House plan. The Complaint further states, “With regard to the African-American communities, the Congressional Plan increases by one 230a the number of congressional districts with a Black Vot- ing Age Population (BVAP) of over 40% and also contains one district with 37.6% BVAP. In addition, there are seven majority-minority districts both in the benchmark map and in the Congressional Plan with a Hispanic Voting Age Population (HVAP) greater than 60% (and one new seat with an HVAP over 50%).” 226. On July 25, 2011, the Court allowed the NAACP to intervene in the consolidated actions. The NAACP Plaintiff-intervenors include the Texas State Conference of NAACP Branches and individually named members and registered voters, including Howard Jefferson, Juanita Wallace, and Rev. Bill Lawson. The Court also allowed three African- American congresspersons—Eddie Bernice Johnson, Sheila Jackson Lee, and Alexander Green—to intervene. Docket no. 67. (The Legislative Black Caucus was also allowed to intervene, but the Caucus voluntarily dismissed its claims on August 2.) 227. On August 31, the Court dismissed Plaintiffs’ 15th Amendment claims. Docket no. 275. On September 2, the Court dismissed Plaintiffs’ partisan gerrymandering claims and claims relating to counting incarcer- ated persons, but denied the motion to dismiss the claims involving census undercount. Docket no. 285. In the same order, the Court also denied Defendants’ motion to dismiss the claims asserted by Morris and the Texas Latino Redistricting Task Force for lack of standing, finding that it was sufficient at the pleading stage for the Task Force to allege that it was asserting claims on behalf of members who suffered an injury- in-fact. From September 6 to 16, 2011, this Court held a bench trial on the claims in this case. 228. On September 29, 2011, this Court enjoined implementation of the House and Congressional plans 231a because they had not been precleared. Docket no. 380. In November 2011, this Court issued interim redistricting plans for the House, Plan H302, and for Con- gress, Plan C220. Docket nos. 528, 544. Texas appealed. 2012 229. On January 20, 2012, the Supreme Court issued Perry v. Perez, 132 S. Ct. 934 (2012), which clarified the governing legal standards and vacated this Court’s interim plans. In February 2012, this Court held a hearing on interim plan issues. On March 19, 2012, this Court entered a second set of interim plans for the Texas House, Plan H309, and Congress, Plan C235, applying the standards in Perry v. Perez. Docket nos. 681, 682, 690 & 691. 230. On August 28, 2012, the D.C. District Court denied preclearance of the State’s enacted plans. Texas v. United States, 887 F. Supp. 2d 133 (D.D.C. 2012) (three-judge court), vacated, 133 S. Ct. 2885 (2013). Texas appealed. 2013 231. The Legislature’s regular 2013 session ended with no redistricting action. But the Governor called the Legislature back for the first called special session to consider “legislation which ratifies and adopts the interim redistricting plans ordered by the federal district court as the permanent plans for districts used to elect members of the Texas House of Repre- sentatives, Texas Senate and United States House of Representatives.” 232. On June 24, 2013, the 83rd Texas Legislature passed new redistricting plans for the Texas House of Representatives and Texas Congressional districts. 232a See S.B. 3 and S.B. 4, 83rd Legislature, 1st Called Session. S.B. 4 “ratified and adopted” this Court’s interim map, Plan C235, without change, and repealed S.B. 4 from the 2011 first special session, which had adopted Plan C185. S.B. 3 adopted Plan H358 as the plan for the Texas House of Representatives and repealed H.B. 150 from the 2011 regular session, which had adopted Plan H283. The new Texas House plan (Plan H358) is largely identical to this Court’s interim plan, but has some changes to Tarrant, Dallas, Harris, and Webb Counties. The plans were sent to Governor Perry for approval on June 24, 2013. 233. On June 25, 2013, the Supreme Court decided Shelby County v. Holder, 133 S. Ct. 2612 (2013), which invalidated the § 5 coverage formula for preclearance in § 4(b) of the VRA. 234. On June 26, Perry signed S.B. 3 and S.B. 4. The plans for the Texas and U.S. House did not become law until September 24, 2013. See TEX. CONST. art. III § 39. 235. On June 27, 2013, the Supreme Court vacated the D.C. District Court’s judgment denying preclearance of the 2011 plans and remanded the case for further consideration in light of Shelby County and the suggestion of mootness of appellees Wendy Davis, et al. Texas v. United States, 133 S. Ct. 2885 (June 27, 2013). The D.C. Court then granted Texas’s motion for voluntary dismissal. 236. On June 28, Defendants filed a motion to dismiss for lack of subject matter jurisdiction, arguing that the case had become moot and should be dismissed. Docket no. 768. This Court held a status conference on July 1, 2013. At the hearing, Plaintiffs expressed a desire to amend their complaints to 233a challenge the 2013 plans, and some Plaintiffs stated their intent to amend their existing claims related to the 2011 plans to seek relief under § 3(c) of the VRA. After the status conference, the Court issued an order summarily denying the motion to dismiss for lack of jurisdiction without prejudice and also issued an order directing Plaintiffs to file motions for leave to amend pleadings by July 12. Docket nos. 771 & 772. Motions for leave to file amended pleadings were then filed. 237. On September 6, 2013, this Court denied Texas’s motion to dismiss claims against the 2011 enacted plans as moot, and allowed Plaintiffs to amend their complaints to assert claims against the 2013 enacted plans and to add claims for relief under § 3(c) of the VRA. Docket no. 886. The Court also ordered that the 2013 plans (Plan C235 and Plan H358) be used as interim plans for the 2014 elections. Docket no. 886. On September 24, 2013, the Court allowed the United States to intervene. Docket no. 904. 2014 238. On June 23, 2014, this Court granted summary judgment on the Fifteenth Amendment claims asserted against the 2013 plans. Docket no. 1108. In July 2014, this Court conducted a six-day trial regarding claims against the 2011 House plan (Plan H283). In August 2014, this Court conducted a six-day bench trial regarding claims against the 2011 Congressional plan (Plan C185). Parties 239. Perez v. State Plaintiffs (docket no. 277 stipulations) Plaintiff Shannon Perez is a Latina citizen and registered voter who resides and is domiciled in Bexar County, Texas. Perez was a registered voter in Pre- cinct 3104 in 2011. PL-127. Plaintiff Harold Dutton, 234a Jr. is an African-American citizen and registered voter who resides and is domiciled in Harris County, Texas and is a member of the Texas Legislature representing the 142nd Legislative District. Dutton was dismissed as a party in December 2011. Docket no. 559. Plaintiff Gregory Tamez is a Latino citizen and registered voter who resides and is domiciled in Bexar County, Texas. Tamez was a registered voter in Precinct 2047 in 2011. PL-130. Plaintiff Sergio Salinas is a Latino citizen and registered voter who resides and is domiciled in Hidalgo County, Texas. Salinas was a registered voter in Precinct 53 in 2011. PL-129. Plaintiff Carmen Rodriguez is a Latina citizen and registered voter who resides and is domiciled in El Paso County, Texas. Plaintiff Rudolfo Ortiz is a Latino citizen and registered voter who resides and is domiciled in Nueces County, Texas. Ortiz was a registered voter in Precinct 79 in 2011. PL-125. Plaintiff Nancy Hall is an African- American citizen and registered voter who resides and is domiciled in Dallas County, Texas. Hall was a registered voter in Precinct 3319 in 2011. PL-128. Plaintiff Dorothy DeBose is an African-American citizen and registered voter who resides and is domiciled in Tarrant County, Texas. DeBose was a registered voter in Precinct 1120 in 2011. PL-126. 240. 240. Latino Task Force v. Perry Plaintiffs (docket no. 277 stipulations) Joey Cardenas is Latino. He is a registered voter who resides in Louise in Wharton County, Texas. Florinda Chavez is Latina. She is a registered voter who resides in Austin in Travis County. Under Plan C185, she lives in CD35. Sergio Coronado is Latino. He is a registered voter who resides in Canutillo in El Paso County, Texas. Under Plan H283, he lives in HD78. Armando Cortez is Latino. He is a registered voter who resides in San Antonio in Bexar County, Texas. Under Plan C185, he 235a lives in CD35. Renato De Los Santos is Latino. He is a registered voter who resides in Dallas in Dallas County, Texas. When this lawsuit was filed, he lived in CD6 in Plan C185. Alex Jimenez is Latino. He is a registered voter who resides in Fort Worth in Tarrant County, Texas. Under Plan C185, he lives in CD12. Emelda Menendez is Latina. She is a registered voter who currently resides in San Antonio in Bexar County, Texas. Jose Olivares is Latino. He is a registered voter who currently resides in Corpus Christi in Nueces County, Texas. Under Plan C185, he lives in CD27. Tomacita Olivares is Latina. She is a registered voter who currently resides in Corpus Christi in Nueces County, Texas. Under Plan C185, she lives in CD27. Alejandro Ortiz is Latino. He is a registered voter who currently resides in San Antonio in Bexar County, Texas. Rebecca Ortiz is Latina. She is a registered voter who currently resides in San Antonio in Bexar County, Texas. Gregorio Palomino is Latino. He is a registered voter who currently resides in San Antonio in Bexar County, Texas. Under Plan C185, he lives in CD35. Socorro Ramos is Latina. She is a registered voter who resides in El Paso in El Paso County, Texas. Under Plan C185, she lives in CD23. Gilberto Torres is Latino. He is a registered voter who resides in Uvalde in Uvalde County, Texas. Under Plan C185, he lives in CD23. Cynthia Valadez is Latina. She is a registered voter who resides in Austin in Travis County, Texas. Under Plan C185, she lives in CD35. Cesar Yevenes is Latino. He is a registered voter who resides in Corpus Christi in Nueces County, Texas. Under Plan C185, he lives in CD27. Under Plan H283, he lives in HD32. 241. Rodriguez v. Perry Plaintiffs (docket nos. 277 & 958 stipulations ) Eddie Rodriguez is a Hispanic and registered voter who resides in Austin, Texas. Under 236a Plan C185, he resides in CD35. Milton Gerard Washington is an African-American registered voter who resides in Austin, Texas. Under Plan C185, he resides in CD10. Bruce Elfant is an Anglo registered voter who resides in Austin. In Plan C185, he resides in CD10. Balakumar Pandian is an Asian-American registered voter who resides in Austin, Texas. Under Plan C185, he resides in CD25. Alex Serna and Sandra Serna are Hispanic registered voters who reside in El Paso, Texas. In Plan C185, they reside in CD16. Betty F. Lopez is a Hispanic registered voter who resides in San Antonio, Texas. Under Plan C185, she resides in CD35. David Gonzalez is a Hispanic registered voter who resides in San Antonio, Texas. Under Plan C185, he resides in CD20. Beatrice Saloma is a Hispanic registered voter who resides in San Antonio, Texas. Under Plan C185, she resides in CD20. Lionor Sorola- Pohlman is a Hispanic registered voter who resides in Houston, Texas. Under Plan C185, he resides in CD2. Eliza Alvarado is a Hispanic registered voter who resides in Pharr, Texas. Under Plan C185, she resides in CD15. Juanita Valdez-Cox is a Hispanic registered voter who resides in Donna, Texas. Under Plan C185, she resides in CD34. Josey Martinez is a Hispanic registered voter who resides in Fort Worth, Texas. Under Plan C185, she resides in CD26. Nina Jo Baker is an African-American registered voter who resides in Fort Worth, Texas. Under Plan C185, she resides in CD26. 242. Travis County, a political subdivision of the State of Texas under Article I, Section 1, of the Texas Constitution, is charged by the Texas Legislature with principal local responsibility for the conduct of elections, including elections for congressional office. Under Section 42.001(a) of the Election Code, the 237a County’s Commissioners Court is responsible for establishing election precincts within its territory and, in redistricting years, is directed in Section 42.032 of the Election Code to complete the process by October 1st. 243. The City of Austin, a political subdivision of the State of Texas, is a home rule municipality under Article XI, Section 5, of the Texas Constitution, with full power of local self-government under Texas Local Government Code § 51.072(a), and is authorized by Article I, § 3, of its charter to take such actions as its governing body deems necessary to advance the interests of its residents. 244. Quesada v. Perry Plaintiffs (docket nos. 277 & 302 stipulations) Under Plan C100, Plaintiffs are citizens and registered voters residing in districts 6, 9, 18, 20, 24, 26, 29, 30, and 33. Plaintiff Margarita Quesada is a Hispanic citizen and a registered voter who resides in San Antonio, Texas. Under Plan C185, she resides in CD20. Plaintiff Romeo Munoz is a Hispanic citizen and a registered voter who resides in Irving, Texas. Under Plan C185, he resides in CD24. Plaintiff Marc Veasey is an African-American citizen and a registered voter who resides in Fort Worth, Texas. Under Plan C185, he resides in CD12. At the time this lawsuit was filed, Marc Veasey was the duly elected representative of HD95. Plaintiff Jane Hamilton is an African-American citizen and a registered voter who resides in Dallas, Texas. Under Plan C185, she resides in CD30. Plaintiff Lyman King is an African-American citizen and a registered voter who resides in Grand Prairie, Texas. Under Plan C185, he resides in CD6. Plaintiff John Jenkins is an African- American citizen and a registered voter who resides in Arlington, Texas. Under Plan C185, he resides in CD33. Plaintiff Kathleen Maria Shaw is an African- 238a American citizen and registered voter who resides in Cedar Hill. Under Plan C185, she resides in CD30. Plaintiff Debbie Allen is an African-American citizen and a registered voter who resides in Houston, Texas. Under Plan C185, she resides in CD18. Plaintiff Jamaal R. Smith is an African-American citizen and a registered voter who resides in Houston, Texas. Under Plan C185, he resides in CD9. Plaintiff Sandra Puente is a Hispanic citizen and registered voter who resides in Houston, Texas. Under Plan C185, she resides in CD29. 245. NAACP and Congresspersons Plaintiff Interve- nors (docket no. 277 stipulations) The Texas-NAACP is the oldest and one of the largest, most significant organizations promoting and protecting the civil rights of African-Americans in Texas. It is a non-profit mem- bership organization with over sixty branches across the state and members in almost every county of Texas. Since its inception, the organization has been involved in numerous voting rights and redistricting cases in the state. Plaintiff Howard Johnson resides in Houston, Harris County, Texas. He is a member of the Texas State Conference of NAACP Branches and is a registered voter. Plaintiff Juanita Wallace resides in Dallas, Dallas County, Texas, and is the President of the Dallas Branch of the NAACP and a registered voter. Under Plan C185, she resides in CD30. Under Plan H283, she resides in HD100. Plaintiff Rev. Bill Lawson resides in Houston, Harris County, Texas and is a member of the NAACP and a registered voter. He is Pastor Emeritus of Wheeler Avenue Baptist Church, and both a long-time civil rights leader and an advocate for the indigent and homeless in Houston. Under Plan C185, he lives in CD18. Under Plan H283, he lives in HD147. 239a 246. Plaintiff Eddie Bernice Johnson is an African- American member of the United States Congress from Dallas representing CD30. She has served in Congress since 1993. 247. Plaintiff Sheila Jackson Lee is an African- American member of the United States Congress from Houston representing CD18, the old district of Barbara Jordan. When this suit was filed, Congress- woman Jackson Lee was in her ninth term in the United States Congress. 248. Plaintiff Alexander Green is an African- American member of the United States Congress from Houston representing CD9. When this suit was filed, Congressman Green was in his fourth term in Congress. Defendants 249. Defendant Rick Perry was the Governor of the State of Texas and chief executive officer of the State of Texas when this suit was filed. The current Gover- nor is Greg Abbott. Defendant Hope Andrade was Texas Secretary of State when this suit was filed. The current Secretary of State is Carlos H. Cascos. The Texas Secretary of State is responsible for administer- ing and supervising the elections of the United States Representatives from the State of Texas. Congress - Houston 250. The benchmark plan had three minority districts in Harris County, one Hispanic and two African- American. CD29 (wholly within Harris County) was a Latino opportunity district (72.3% HVAP, 56% HCVAP using 2005-2009 ACS data), and was represented by Gene Green (Anglo, Democrat). CD9 (in Harris and Fort Bend Counties) and CD18 (wholly within Harris 240a County) were considered to be African-American districts that consistently elected minority candidates of choice. CD9 was 38.9% HVAP, 36.3% BVAP, 48.2% Black alone CVAP, and 48.4% combined BCVAP, and was represented by Al Green (African American, Dem- ocrat). CD18 was 39% HVAP, 37.9% BVAP, 46.4% Black alone CVAP, and 46.7% combined BCVAP, and has been represented by Sheila Jackson Lee (African American, Democrat) since 1994. 251. The November 20, 2010 joint subcommittee field hearing in Houston had a large turnout. D-588 at 8. Members of the committee noted the large minority growth in Harris County. Congressman Gene Green (CD29) testified, noting the minority growth and stating his hope that the Legislature would add another majority-Hispanic congressional district. Id. at 13. Congressman Al Green (CD9) also noted the large minority growth. Id. at 19-20. George Korbel testified that the minority growth was responsible for Texas gaining the four new congressional seats. Id. at 33. He noted that there were seven congressional districts that touched Harris County and that Hispanics were 45% of the population of Harris County, yet no His- panics were elected to Congress from Harris County (CD29 was a Latino opportunity district but it was represented by an Anglo Democrat). Id. at 34. He testified that Houston deserved another Hispanic congressional seat. Id. at 37. Herlinda Garcia and Mary Almendarez of LULAC testified that the Hispanic growth justified another Hispanic district in Houston. Id. at 58, 74-75. Congresswoman Sheila Jackson Lee (CD18) noted that her district originated in 1973 and had been represented by Barbara Jordan. She described the core neighborhoods of her district, noted her hope that communities of interest would be respected, and stated that she wanted to work with the 241a Legislature on redistricting. Id. at 24. Residents of CD18 asked that the district and their communities be kept intact. Id. at 57-58. 252. Harris County had approximately 25.36% HCVAP in 2010. TrA1828 (Alford); D-231. Harris and Fort Bend Counties added about 920,000 people between 2000 and 2010, and all of that growth was minority (the Anglo population declined by about 42,000). Joint Expert Ex. E-4 (Murray report) at 27. Hispanic population grew by 615,885 and African-American population grew by 209,375 (a combined growth of 824,260). Id. Murray noted that the minority growth tended to be in the same areas (the eastern half of Fort Bend County and southwest Harris County), and opined that it would be easy to create a second district that would give Latino voters an opportunity to elect while maintaining African-American opportunity districts CD9 and CD18. Id. Downton testified that he tried to draw a new Latino opportunity district in Harris County (using a 50% HCVAP standard), but could not because CD29 has a non-suspense SSVR of 52.6% and a 56% HCVAP, and it was not possible to draw a new district and keep CD29 above 50%. Downton 8-12-11 depo. at 73. 253. The first public plan, Plan C125 drawn by Downton and released May 31, did not add any new minority districts in the Houston area. Districts 9, 18, and 29 remained in the same general area but CD18 lost population in central and north Harris County and extended into southwest Harris County for the first time. CD18 was 48% combined BCVAP using 2005-2009 ACS data. CD29 moved out of eastern Harris County and gained new population in central Harris County that had been in CD18. CD29 was 54.8% HCVAP using 2005-2009 ACS data. CD9 was 242a significantly reconfigured, taking in more of Fort Bend County. It was 47.6% combined BCVAP using 2005- 2009 ACS data. New CD36 (referred to as the “horseshoe” or “jumbo shrimp” district) began in Harris County and then extended in a horseshoe shape through numerous counties to the north and east. 254. Congresswoman Sheila Jackson Lee (CD18) issued a statement on June 2 opposing Plan C125. She was disappointed that the plan did not reflect the large minority growth, and stated that in her district, CD18, “historic neighborhoods are divided and major communities of interest have been carved from the district.” D-664. 255. At the June 2 HRC hearing on Plan C125, Rep. Walle (Hispanic, Democrat) and Rey Guerra of the Greater Houston Civic Coalition complained that no new Hispanic district had been created in the Houston area. D-601 at 42-43, 115-16. Walle also noted that minority voters in Harris County act in coalition. Id. at 43. Rep. Alonzo (Hispanic, Democrat) agreed with the need for a new Latino opportunity district in Harris County. Id. at 50. Rep. Alvarado (Hispanic, Democrat) noted that testimony from the Houston field hearings was not reflected in the plan. Id. at 87. She submitted Jackson Lee’s statement opposing Plan C125 into the record, and noted that Congressmen Al Green and Gene Green opposed the map. Id. at 21. MALDEF proposed Plan C122 and C123 with a new minority coalition district in Harris County. Id. at 79. In response to complaints about the new CD36 horseshoe district, Solomons stated that there might be changes. Id. at 67-68. 256. Solomons’ statewide substitute Plan C130, made public on June 2, included changes in Harris County, including minor changes to CD18 and CD29, 243a and changes to CD9, CD7, CD10, CD36 (to get rid of the horseshoe), and CD2. CD36 was now located in southeast Harris County and joined with numerous southeastern counties. 257. At the June 3 hearing before the SRC, Joey Cardenas of LULAC noted that minority groups’ maps proposing a district in Harris County had been available during the regular session and there should have been discussions about them before, rather than “at the last minute, in a Special Session.” D-602 at 110. Sen. Gallegos presented a four-district demonstration map, C127, that created four minority districts in Harris County. Id. at 175-76. Plan C127 maintained the existing three minority districts and added CD36 in southeast Harris County. He also proposed demonstration map C129, which created three minority districts within Harris County and another district joining part of Harris County with part of Fort Bend County and parts of four other counties along the coast. Gallegos also presented a statewide substitute (likely Plan C131, with a similar Harris County configuration as Plan C127) and moved for adoption, which was not acceptable to the author. Id. at 178. It did not pass (8 to 6). Id. at 179. Nina Perales of MALDEF proposed Plans C122 and C123 again. Id. Seliger laid out Plan C130, stating that it got rid of the horseshoe and made CD36 a southeast Texas district, made Jefferson and Orange Counties whole, and moved CD2 entirely into Harris County and made it more compact. Id. at 180. 258. On June 6, Opiela sent Downton an email titled C130-Houston Fix with an attachment called “Plan 130 Revision Suggestion” and wrote, “What I talked to you on the phone about Friday.” PL-311 Pt. 7 at 445. 244a 259. On June 9 at the HRC formal meeting, Solomons laid out Plan C144, which exchanged some population between CD18 and CD29 in downtown Houston and north Houston and also some population between CD8 and CD10 in Harris County to reunite the City of Tomball in CD10 (from C141, the Senate engrossment plan). Alvarado proposed statewide Plan C126,which had CD9 (36.9% BVAP), CD18 (39.2% BVAP), CD29 (62.4% HVAP), and CD36 (62.3% HVAP).14 260. On June 14 during the House floor debate, Rep. Alvarado (Hispanic, Democrat) offered Plan C168 for southeast Texas, which she said would add a second Latino opportunity district in Harris County while maintaining the three existing districts. She noted that both districts CD29 and CD36 were completely in Harris County and had HVAP over 50%, with the SSVR slightly under 50%, but “both have proven to elect Latino voters’ preferred candidate.” D-23 at S7. She said, “As the Department of Justice guidelines clearly state, there is no specific demographic percentage, no magic number that determines effectiveness. It’s a function of different factors such as turnout and election results. This map does not pair any incumbents. District 29, which is the current Latino opportunity district in Harris County, remains anchored in north Houston and picks up the growing Latino communities of west and southwest Houston, Harris County. Districts 9 and 18 remain as coalition districts that currently would like to remain represented by African American congress members.” Martinez Fischer supported the amendment because

14 The Court lacks CVAP data for this plan. Plan C168 has a very similar but not identical configuration of these districts, but the Court lacks CVAP data for Plan C168 as well. 245a it added a new minority district in Harris County, and “it demonstrates . . . something that we will be doing later with some subsequent statewide amendments.” Id. at S8. Solomons opposed the amendment, stating that the map retrogressed CD29 because it dropped the SSVR to 35.5% and its HCVAP to only 38.6%, and the new district was only 42.5% SSVR and 41.1% HCVAP, and that it in effect had one less Latino opportunity district. Id. at S9-S10. Alvarado responded that they were coalition districts. Veasey stated that Solomons was using arbitrary numbers that “are not the gold standard” and that there are other factors that need to be taken into consideration, including that the proposed district would elect the candidate of choice. Id. at S10. Solomons moved to table the amendment, which passed 94 to 47. Id.; PL-218 at 3-4. 261. Rep. Martinez Fischer offered Plans C163 and C164, but Solomons said the new proposed CD36 in Harris County was not a majority-Hispanic district because it failed to reach 50%. D-23 at S23. Rep. Dukes (African American, Democrat) offered Plan C166. She said that the Harris County configuration was similar to the plan presented by Representative Alvarado, and the new CD36 had the same Hispanic performance and slightly higher black performance. Id. at S30. Using 2005-2009 ACS data, Plan C166 had 7 HCVAP- majority districts statewide. In Harris County, CD9 was 40.5% BCVAP, CD18 was 44.3% BCVAP, CD29 was 38% HCVAP + 21.2% BCVAP (a coalition district), and CD36 was 45.1% HCVAP (and 14.2% BCVAP). Joint Map Ex. J-7. Solomons opposed, stating, “I have some legal concerns about the map. It creates one less Hispanic district – Hispanic majority district – than the committee map, and neither the new District 35 in North Texas nor District 36 in Harris County are Hispanic minority districts. In fact, this map also 246a retrogresses District 29, and quite frankly Ms. Dukes[‘] map, as I say, creates one less majority district than the committee map.” D-23 at S36. He noted that in the benchmark CD29 has an SSVR of 52.6% and an HCVAP of 56%, but in C166, the SSVR drops to 34.6% and HCVAP drops to 38%. Id. Dukes responded, “I’m not quite certain where Representative Solomons pulls his numbers and sometimes I wonder about the concern put forth on minority communities, when Con- gressional District 29 is used as the litmus test for what is considered retrogression when one arguably can state something different.” Id. at S37. The amendment was tabled. Id. at S39. Rep. Alonzo entered an exhibit that complained that no new minority district was created in Harris County. Id. at S-54-S57. 262. Plan C185 in Harris County included the same minority districts as the benchmark: CD9 (in Harris County and Fort Bend County) was 47.5% BCVAP and 18.3% HCVAP (22.6% Anglo CVAP); CD18 (entirely within Harris County) was 48.6% BCVAP and 17.4% HCVAP (29.4% Anglo CVAP); CD29 (entirely within Harris County) was 56.3% HCVAP. D-656 shows the VTDs at least 50% Hispanic or African American CVAP in the Houston area in C185, and it shows that most of the 50% HCVAP VTDs are in CD29, and most of the 50% BCVAP VTDs are in CD9 and CD18. Dr. Alford testified that there was no evidence that the Legislature could create an additional HCVAP- majority district in Harris County and he had not seen such a map. TrA1847. CD2 and CD7 are completely within Harris County, and are majority-Anglo CVAP (66% and 68.3%). CD2 and CD7 both join central Harris County minority population with Anglo population within Harris County. CD2 begins in central Harris County and wraps around minority districts 18 and 29 and picks up Anglo population in northeast 247a Harris County. LULAC-4-D. CD7 combines significant minority populations with predominantly Anglo areas within Harris County. LULAC-16-J. CD2 has 214,632 Hispanic persons, and CD7 has 202,206 Hispanic persons. Five other districts come into Harris County: CD8, which joins portions of northern Harris County with counties to the north; CD10, which joins portions of western Harris County with portions of Travis County; CD22, which joins portions of southeast Harris County with parts of Fort Bend and Brazoria Counties; CD36, which joins parts of eastern Harris County with southeast Texas counties; and CD14, which joins a small portion of southeast Harris County with portions of coastal counties (Brazoria, Galveston, and Jefferson). 263. Korbel opined that Plan C185 utilizes cracking and packing to avoid creating any new minority districts. Joint Expert Ex. E-11 at 6. Dr. Arrington testified that, instead of creating a new minority district, there are number of Anglo-dominated suburban districts with fingers reaching in and cracking Hispanic population, making it impossible to create an additional district in Harris County. TrA413. Areas of high minority population in eastern Harris County (including 142,385 Hispanic persons) are joined with mostly- Anglo Liberty, Polk, Hardin, Tyler, Jasper, Orange, and Newton Counties in CD36. LULAC-4-C, LULAC- 16-K; US-352 (Arrington report) at 49 Table 15. CD10 joins Harris County minority population in western Harris County (including 62,822 Hispanic persons) with minority population in northeastern Travis County, connected by miles of Anglo-majority rural counties. LULAC-4-A, LULAC-16-I; US-352 (Arring- ton report) at 49 Table 15. Arrington concluded that 672,362 Hispanic persons were available in Harris County to create an additional Hispanic district, but 248a they were cracked instead. TrA412-13;US-352 (report) at 48-49. 264. Murray testified that benchmark CD9, represented by Al Green, was almost exactly the right population (only 34,000 over), was reasonably compact, and was effective, so it could easily have been adjusted for voting rights and practical political traditional principles with modest modification, such as by moving six or so precincts to next-door CD29, which was a little underpopulated. Tr1045, TrA1387; Joint Expert Ex. E-4 at 28. Instead, CD9 was substantially and unnecessarily changed, without any consultation with Green. TrA1387-88 (Murray). Murray testified that Plan C185 pretty much dismantles it and rebuilds it, moving a couple hundred thousand people around, taking out the core, including Al Green’s office, and also removing important assets like the Medical Cen- ter that he has an important constituent relationship with, which did not follow the principle of starting with existing districts if they do not need much modification. Tr1046, TrA1392. Murray testified that Green’s political base was in the older, inner-city precincts of Harris County, but a lot of new suburban and not-yet-fully-developed areas were added, which creates unpredictable new growth in what had been a mature district, and creates the potential for a tension district or one that loses its character as an African- American district. Tr1046, Tr1050, TrA1392. The new portions of Fort Bend County added in are much less African American and much more Asian and Hispanic, which creates a potentially very different district. Tr1047 (Murray). Enacted CD9 is 38.3% Black and 38.8% Hispanic in terms of total population. Tr1059 (Murray). This creates a lot of potential for tension, particularly if there is no other opportunity district for Hispanics in the Houston area given the explosive 249a growth. Tr1048 (Murray). Murray testified that there is strong evidence of African-American and Hispanic coalition voting, but also stated that without better opportunities for a growing Latino population as we move through the decade, there will be some real tensions. Tr1061. He agreed that, in the primary, African Americans tend to vote for African Americans, and Hispanics for Hispanics. Tr1061. He testified that in demonstration Plan C193, the population of CD9 would be 40.3% Black and 37.3% Hispanic, restoring a Black plurality, and the district would not include so much of the growth potential areas, making this map substantially more likely to maintain African- American opportunity in CD9. Tr1049. 265. Murray testified that benchmark CD18 was overpopulated by about 24,000, so again only minor changes were required, but major changes were made. Tr1051; Joint Expert Ex. E-4 at 29-30. He noted that Plan C185 split traditional important neighborhoods in Houston like the Third Ward in the MacGregor area, which is home to many of the African-American opinion leaders in Houston and had been in CD18 since its creation but was divided between CD9 and CD18. He further testified that important areas like downtown Houston were removed, which have been in this district since Barbara Jordan represented it in the 1970s. Tr1051. He also testified that the district was extended into southwest areas with no connection to the existing district. Murray testified that there was major unnecessary surgery that negatively impacted CD18 and the member-constituent relationships there. Id. He felt that although it was a performing African- American district as drawn, it had the potential to become a tension district. TrA1394-95 (Murray). He also felt that the changes unnecessarily created potential tension, and that benchmark CD18 was much less 250a likely to lead to tension than the new C185 version. Tr1073. He testified that the NAACP Plan C193 restores the original district with only minor modifica- tions. Tr1052. 266. Dr. Ansolabahere noted that Harris County’s population is distributed across ten congressional districts (2, 7, 8, 9, 10, 14, 18, 22, 29, and 36), and Harris County residents are a majority or plurality in seven of those districts. Joint Expert Ex. E-15 at 20. Harris County is 41% Hispanic, 33% White non- Hispanic, and 19% Black. The three districts in which Harris County is not a majority or plurality are all White-majority districts. Id. at 21. Of the seven districts weighted in Harris County, only one is majority Hispanic and none are plurality Hispanic, even though Hispanics are the single largest racial/ethnic group in the area with 1.7 million persons (more than two times 698,488, the population of a district). Id. There are two Black-plurality districts and no Black- majority district. Even though Whites are only one- third of the Harris County population, they are an outright majority in four districts. Id. Looking at distribution, the Hispanic population is distributed in such a way that only 31% of Hispanics are in a Hispanic-majority or plurality district. Although they are the largest racial/ethnic group in the county, they are distributed in such a way that they have the least representation. Id. Two-thirds of the Black population is in the two Black-plurality districts. Over 80% of Whites are in districts that are majority-White. Id. at 21-22. 267. Despite the minority growth in Harris County/ Houston, no new minority districts were created. Instead, the existing three minority districts were maintained as minority districts (though they were 251a significantly altered). CD9 (in Harris County and Fort Bend County) was 47.5% BCVAP and 18.3% HCVAP (22.6% Anglo CVAP); CD18 (entirely within Harris County) was 48.6% BCVAP and 17.4% HCVAP (29.4% Anglo CVAP); CD29 (entirely within Harris County) was 56.3% HCVAP. The remaining minority population is cracked among Anglo-majority districts where minority voters will not be effective. 268. Plaintiffs propose maps with an additional Hispanic district, but no proposed maps contain two HCVAP-majority districts using 2005-2009 ACS data. No proposed maps create an additional BCVAP- majority district. Rather, Plaintiffs’ proposed maps include a new coalition district. Congress - DFW area 269. In benchmark Plan C100, six districts were wholly (CD30 and CD32) or partly (CD3, CD5, CD24, CD26) in Dallas County and four districts (CD6, CD12, CD24, CD26) were partly in Tarrant County (CD24 and CD26 were in both counties, such that eight districts were wholly or partly in both counties). CD30 was an African-American opportunity district; it was 49.8% Black Alone CVAP and 50.3% combined BCVAP (Black alone plus mixed Black+White plus mixed Black+American Indian) using 2005-2009 ACS data. It was the only minority opportunity district in the DFW area, and the incumbent was African- American Democrat Eddie Bernice Johnson. All the other districts were represented by Anglo Republican incumbents. 270. CD26 in the benchmark configuration contained a majority of Denton County, and had a northern extension going up to the top of Cooke County and a southern extension going down through Tarrant 252a County/Fort Worth, including the cities of Keller, Richland Hills, North Richland Hills, and Hurst. TrA1607 (Downton); TrA458 (Arrington). The southern extension was not predominantly Hispanic. Tr942 (Downton). 271. Overall population growth slowed in Dallas after 2000, largely due to the out-migration of Anglos, whose population declined. Joint Expert Ex. E-4 (Murray report) at 30. Tarrant County grew faster, with almost all growth being minority. Id. at 31; TrA1173-74 (Moss). Fort Worth is majority minority in total population. TrA1173 (Moss). Although DFW and Houston have similar populations, in the benchmark plan DFW had only one minority congressional district, while Houston had three. Joint Expert Ex. E-4 (Murray report) at 32. 272. In Dallas County in 2010, there were approximately 1,360,390 citizens of voting age, 649,069 of whom were Anglo, 277,395 of whom were Hispanic, and 359,125 of whom were African American. TrA963 (Ansolabahere); Rod-912 at 44; D-181. Dallas County was approximately 20.39% HCVAP. TrA1828 (Alford); D-231. In Tarrant County in 2010, there were approximately 1,141,750 citizens of voting age, 730,125 of whom were Anglo, 176,280 of whom were Hispanic, and 175,795 of whom were African American. TrA963 (Ansolabahere); Rod-912 at 44; D-181. Tarrant County was approximately 15.44% HCVAP. TrA1828 (Alford): D-231. Thus, minorities were 47% of the CVAP of Dallas County, 31% of the CVAP of Tarrant County, and 39.5% of the combined Dallas and Tarrant County CVAP. Yet they had only one opportunity district (CD30, an African-American district) out of the eight districts (12.5%) that came into those counties. To 253a explain the lack of a Latino district, Dr. Alford testified that the combined HCVAP of approximately 450,000 was substantially lower than the HCVAP of Bexar (593,520) and Harris (590,280) Counties. TrA1842. 273. During the regular session (early April), the congressional delegation submitted a proposed map that created a new minority district in DFW by combining minority populations in Dallas and Tarrant Counties. The delegation stated that they wanted to add a new DFW VRA district for two reasons: (1) to minimize the possibility of a successful § 2 claim since there are over a million voting age Hispanics in DFW, and they do not have either an effective coalition or Hispanic-majority district and (2) politically, if a new Democratic district is not created in DFW, the long-term prospect of Congressman Sessions’ district (CD32) and, to some extent, Congressman Marchant’s district (CD24) as Republican-majority districts becomes unlikely in the out years of the decade. D-611; PL- 1025. Congresswoman Johnson supported the new DFW district and testified that Black and Latino voters had worked in coalition in Dallas County. Tr1277, 1283. 274. Opiela emailed Downton a delegation map on April 6, and it had the new proposed minority district CD33 in DFW. TrA1697 (Downton); PL-587. Downton imported the map into his RedAppl account, and it appears on April 6 as hrc1C104 (“Congressional 5”). CD33’s HVAP was 61.3% and HCVAP was 40.2% using 2005-2009 ACS data. TrA1698 (Downton); PL-587. The new district put Hispanic neighborhoods from Dallas and Tarrant Counties together and encompassed primarily urban and suburban areas. TrA1368-69 (Solomons). In the delegation map, there 254a was only a small protrusion from CD26 (the Denton County-based district) into Tarrant County. Downton testified that this “made sense” because “most of the Democratic votes within Tarrant and Dallas County are absorbed into” the minority coalition district, and the mapdrawer would not “have to worry about balancing the Republican districts” (i.e., splitting up the Democrat votes equally into the surrounding Republican districts) because they would all be heavily Republican. TrA1698. Proposed CD33 had an area rubber band score of .439 and a perimeter to area score of .064. PL-587. Interiano thought proposed CD33 in DFW was compact, but it was not above 50% SSVR, which was the population metric he focused on for Hispanic opportunity districts. TrA382; PL-587. 275. The HRC held a hearing on April 7, 2011 (before a map was released) for invited and public testimony on potential Texas congressional districts in light of the 2010 Census data. D-116 at 111; D-594. Rep. Alonzo (Hispanic, Democrat) advocated for a minority district in the DFW area. Tarrant County Commissioner Roy Brooks testified that the African- American community of southeast Fort Worth had been harmed in the 2003 redistricting by being placed as an appendage to Denton County in CD26, and asked that the community be reinstated as the core of a district made up of Tarrant and Dallas County voters. Id. at 31-36. Sergio DeLeon and others advocated for a Latino opportunity district in North Texas/DFW, and Rep. Peña (Hispanic, Republican) expressed his support for such a district. Id. at 60. Peña stated that creation of such a district was “compelling” given the minority growth. Id. at 60-61. Nina Perales of MALDEF proposed maps C108 and C109 from the Texas Latino Redistricting Task Force. Id. at 75. According to Perales, both maps contained nine Latino 255a opportunity districts, including a new district in DFW (these maps have not been provided to the Court). Id. at 77-78. She stated that over 1.3 million Latinos lived in the metroplex but there was no Latino-majority district in which they could elect their candidate of choice to Congress, that there were enough Latinos in the metroplex to compose the majority of a district, and that the remaining factors (racially polarized voting, history of discrimination) were present under the VRA to warrant the creation of the district. Id. at 78-80. Perales asserted that there was unity between Latinos in the DFW metroplex to create a district. Id. at 84. MALDEF also submitted Plans C122 and C123 during the regular session that would create a new minority district (CD35) in DFW by joining minority populations in Dallas and Tarrant Counties. CD35 was 45% HCVAP using 2005-2009 ACS data and 45.9% using 2008-2012 ACS data. Joint Map Ex. J-3; D-546.6. Proposed CD35 had an area rubber band score of .302 and a perimeter-to-area score of .038. Lay witness Alex Jimenez testified that the western portion of the district in Fort Worth shares common interests with similar neighborhoods in Dallas. PL- 457 (Decl. of Jimenez in support of interim plan C213’s CD34, which is the same as CD35 in Plans C122 and C123). 276. Downton testified that he tried to draw a Latino opportunity district in the metroplex in mid-to- late April, but was unsuccessful because he was not able to draw a district that was majority-HCVAP or SSVR because the Latino population was too spread out and there were too many non-citizens. Tr906-07, Tr975, TrA1795-96; Downton 8-12-11 depo. at 68, 127. He used Hispanic shading when doing this. Tr913, Tr976 (Downton). 256a 277. In early May, the Governor’s office also submitted a proposed map with a new minority district in DFW (CD33). TrA1700, TrA1761 (Downton); TrA1365- 67 (Solomons). The map appears in Downton’s RedAppl account as hrc1C113, created May 2, 2011. TrA1700 (Downton); PL-744. The HVAP of CD33 was 51.6% and the combined H+BVAP was 67%. TrA1701 (Downton). The HCVAP using 2005-2009 ACS data was 33%. PL-744. Downton made revisions to the proposed district to try to get the Hispanic numbers up (shown in hrc1C114), but could not get to 50% HCVAP. TrA1701-02 (Downton); PL-745 (HCVAP is 38.2%; HVAP is 58.5%). Downton used the HCVAP data from the TLC, which was the 2005-2009 ACS survey data. Tr981 (Downton). Due to the ACS data lag, Hispanics and Blacks are a higher share of CVAP in Dallas and Tarrant Counties than the 2005-2009 ACS indicates. Joint Expert Ex. E-15 (Ansolabahere rebuttal report) at 13. Downton did not consider the lagging nature of the ACS survey. Tr982 (Downton). 278. Hanna did some test mapping to see if he could draw a majority-Hispanic district in DFW. TrA1576 (Hanna). He had seen a plan with 50.1% HCVAP population, but thought it had a lot of arms and tentacles, so he tried to draw something more compact, but he could not. TrA1576-77 (Hanna). He did not recall sharing his conclusions with any members. TrA1577. 279. In May 2011, Congressman Lamar Smith and the Republican congressional delegation were continuing to advocate to legislative leadership in favor of creating a new VRA district in the DFW area. TrA246 (Seliger), TrA1703-04 (Downton); PL-311; PL-1114 at 203. Smith’s proposal asserted that it created three Republican seats and a Democrat seat (a 3-1 split), but Downton thought it was a 2-2 split because it would 257a elect two more Republicans and two more Democrats than the benchmark map. TrA1705-07 (Downton). Downton was aware of the potential benefits for Republican DFW delegation members from creating a minority district in DFW in terms of making their districts more Republican. TrA1703 (Downton). 280. Downton testified that he had not seen any proposed maps with a majority-HCVAP or SSVR district in DFW. Downton 8-12-11 depo. at 68, TrA1605. Maps he received from MALC, MALDEF, the Republi- can Congressional Delegation, and Rep. Veasey were not at that 50% level. Tr908, TrA1795-96. Downton testified that he looked at the maps and tried to tweak them by going to the block level and seeing if he could bump the number up, but he was not able to. Downton 8-12-11 depo. at 126-27; Tr941. He testified that he could not get to 50% SSVR (he knew CVAP is going to be higher than SSVR), and thus he did not feel the district was required. Tr906, Tr981-82; Downton 8-12- 11 depo at 68. Downton never analyzed the performance of the proposed minority DFW districts. TrA1798 (Downton). Korbel testified that proposed CD33 in the Republican delegation plan would elect the Hispanic candidate of choice based on the county commissioner precinct elections. Tr718. 281. Korbel agreed that he had no demonstration map showing SSVR or HCVAP above 50%. Tr741. Korbel testified that LULAC’s demonstration plan C195 included a district that was designed to come up with an HCVAP of 50%. Tr736; LULAC-12. It connects the minority community in Dallas and Tarrant Coun- ties. The district is similar to the Republican delegation proposal’s district, but “more stressed” to raise the HCVAP. Using 2005-2009 ACS data, it is 47.5% HCVAP, and the total SSVR is 44.2%. Tr738 (Korbel). 258a Korbel questioned the accuracy of the ACS data, stating “I can’t imagine a situation where . . . the registered voters would be that close to the citizen voting age population in a developing area like Tarrant County.” Id. He noted that the margin or error for the ACS data was 1%, and that the data was three or four years old, such that he would use the SSVR and project the CVAP from there, and he believed it would reach majority status. Tr738-41. 282. The Task Force offers Plan C190 (Joint Map Ex. J-11), which creates an HCVAP-majority district (CD6) in Dallas and Tarrant Counties using 2005- 2009 ACS data. CD6 is 66.8% HVAP, 50.4% HCVAP (using 2005-2009 ACS data); and 41.6% total SSVR. It has an area rubber band score of .303 and a perimeter- to-area score of .018. Alex Jimenez testified that proposed CD6 joins people, businesses, and consumer bases in those two areas that are fairly similar. Tr574- 75. He thinks voters in the two areas would support the same candidate based on what the candidate represents, not on whether the candidate is Republi- can or Democrat. Tr576. 283. Downton said that although he sought input from other sources, Solomons was the only person directing him. TrA1797. Downton testified that Solomons saw the various proposals. TrA1796. Downton testified that TLC attorneys agreed with him that a DFW Latino district was not a required district, because it could not get to 50%, and Downton told Solomons. TrA1797; TrA1301 (Solomons). Downton testified that he received approval from Solomons to not include a DFW minority district in mid-to-late May. TrA1772. Solomons testified that he and Seliger made the decision because they were told that the numbers were not there and it was not legally required. TrA1300, 259a TrA1316. Seliger testified that he, Doug Davis, Interiano, and Downton discussed whether they were required to create a minority district in DFW. TrA273. It was possible to draw a minority-majority district that combined Hispanics and African Americans in DFW, but Seliger concluded that such a district was not required by the VRA. TrA247 (Seliger). He concluded that drawing an HCVAP-majority district in DFW would lead to an unusually shaped district. TrA246. 284. Solomons stated that the Hispanic population had too many noncitizens and non-voting age persons, and he did not recall seeing any proposals with 50% HCVAP. TrA1301, TrA1323. He said that the Republi- can congressional delegation was drawing a coalition district, and he did not believe those were required. TrA1324. Solomons testified that, if he was told it was not legally required, it meant to him that they could not do it. TrA1301-02. He said it would have been very difficult to get the plan through because there were 101 Republicans and none of the Democrats were voting for a map, so he had to rely solely on Republican votes in a partisan House. TrA1323-24. He said the decision to move forward with a map that did not have the minority DFW district was “disap- pointing.” TrA1300. However, Solomons understood a new majority-minority district in north Texas could be permissible even if it was not legally required. TrA1302 (Solomons). Solomons also understood that creating the district might have a double benefit of creating a Hispanic district and strengthening the Republican performance in neighboring districts TrA1363-64 (Solomons). But he said it would be hard to argue that he was doing it because he thought they should even though it was not legally required. TrA1323-24. However, he acknowledged that, given 260a the Governor’s proposal and the Republican delegation proposal, he could have made an effort to persuade his fellow Republicans in the House that it was a good idea to create this majority-minority district from a Repub- lican perspective, but he did not. TrA1367 (Solomons). He also stated (incorrectly) that he and his staff were “not the lead players” on congressional redistricting, and if the Senate side had wanted to include the district, they would have dealt with it and it would have been fine with him. TrA1302. Yet Seliger said that he deferred to Solomons and his staff in DFW because Solomons is from that area. TrA246. And Downton testified that Solomons never told him that he wanted a minority opportunity district to be created in the DFW area. TrA1786. 285. Because a minority-coalition district in DFW would be a Democrat district, mapdrawers and legislative leaders refused to draw such a district unless they determined that it was compelled by the VRA. Tr906- 07 (Downton); TrA381 (Interiano). Solomons said that if the proposed DFW minority district was not a required district, he could not pass a map that was not a 3-1 map. TrA1599-1601, TrA1796 (Downton); TrA1322- 24 (Solomons). Downton did not recall personally trying to draw a H+BCVAP-majority district, but recalled seeing maps that did that. Tr1011-12. He agreed that such a combined district would likely be over 50%. Tr1011. 286. By mid-May, Downton was drawing the congressional map without a new Latino opportunity district in the DFW area. TrA1772 (Downton). After the decision was made not to include a new minority district, Solomons directed his staff to figure out how to draw the districts based on the delegation members’ input. TrA1303 (Solomons). Downton testified that 261a CD24 was one of the first districts he drew because it is Congressman Marchant’s district, and he is Solomons’ congressman. Tr981. Downton testified that he considered the political function of how to balance the districts, and he “honestly didn’t think about ethnic shading at that point, because it was a political decision.” Tr913. However, he was aware of where the Hispanic population in DFW was located because he considered race when trying to draw a Hispanic- majority district. Tr977-78. Downton still needed to add a district in north Texas because of population growth there, so he put new district CD33 in Tarrant, Parker, and Wise Counties. He stated that he had to balance the percentage of Republicans across the north Texas districts to be within a few percentage points of each other; he tried to draw all the districts in Tarrant County to be at least 55% McCain but below 60%. TrA1606, TrA1626, TrA1793 (Downton). Without a new minority district, it was much harder to balance all of the districts to where the congressmen felt that they still had strong Republican districts. 287. Plan hrc1C109 (US-731F) from late April contains the same proposed minority coalition district from a map provided by Opiela on behalf of the delegation (hrc1C104). Plan hrc1C116 (US-731G) (“labeled Congressional 5/3”) still contains a possible minority coalition district (CD33), though modified somewhat from Opiela’s district (Downton stated that he modified the delegation’s district to try to get it above 50% HCVAP), and then hrc1C117 (US-731H) and hrc1C119 (US-731J) returns to a configuration substantially similar to the delegation’s proposal. By hrc1C124 (US- 731P), however, DFW begins to take its final shape. Plan hrc1C124 was created May 11 and last saved May 20. In it, the minority coalition district is removed, now divided up (cracked) primarily among CD6, CD12, 262a CD26, and CD33. The CD26 extension from Denton County is extended and now includes a significant portion of what had been in the minority district. CD12 and CD33 then take the other portions of what had been in the minority district in Tarrant County, and CD6 shifts east into Dallas County to take a large portion of the former proposed minority district within that county. CD24 remains largely the same, as do portions of the border between CD30 and CD32, indicating that this was not a new map drawn from scratch, but a modification of the prior map that essentially dismantled the minority district. This contra- dicts Defendants’ assertion that once Downton determined that he could not draw a 50%-HCVAP district, he simply drew districts based on partisan criteria and balancing districts between Republicans and Demo- crats. The progression of maps shows that he included a minority district in the map, then cracked it into Anglo-majority districts. 288. In Plan hrc1C130 (US-731-P) (labeled “Con- gressional Ryan Merge 5/20”), CD26 maintains the long extension into Tarrant County that includes the City of North Richland Hills. D-709. 15 In Plan hrc1C131 (US-731Q) (created May 23), the CD26 extension shifts to the west and begins to take the shape found in the first public plan, Plan C125. PL- 1135. The extension now only takes in a small portion of the benchmark CD26 population in Tarrant County.

15 Downton testified that while working on hrc1C130 in RedAppl, he would display population deviation numbers and McCain-Obama statistics (as a % for McCain), as well as partisan shading at the precinct level. TrA1612. He denied using racial shading when working on this map or hrc1C131. TrA1613, 1617. However, at this point he had already cracked the minority district. 263a D-710. Downton testified, “I had it originally closer to where it was in the benchmark, then it moved west when I had to put North Richland Hills . . . in 12.” TrA1709. Downton testified that he had a conversation with Chairman Geren, whose congresswoman was Kay Granger, and Geren said that North Richland Hills had to be a part of Granger’s district, CD12. TrA1614. Downton said he drew at the precinct level along the city lines of Haltom City and Richland Hills and then West Lake and Keller to include them in CD12. TrA1614-15. Downton stated that he was also trying to the put the Trinity River project into CD12 based on where Geren said it was. TrA1615-17. Solomons did not instruct Downton to make these changes and only knew about the Trinity River Project because Downton told him. TrA1370-71 (Solomons). 289. After Downton finished working on hrc1C131 (likely between May 25 and May 28), Downton met with Interiano and Doug Davis, and Interiano told Downton that he split the Fort Worth black population in Tarrant County and asked if he could fix it. TrA1618, TrA1622, TrA1783 (Downton). On May 28, Downton emailed Interiano and Doug Davis stating, “Changes made to keep the black population together in District 12.” US-630. Downton testified that he would have pulled up racial shading at the VTD level to determine that he had “segmented the black population of Fort Worth into a bunch of different districts.” TrA1619-20. Downton testified that he made the change because they were “trying to keep communities of interest together, when possible” and that was one of the redistricting principles they were to consider. TrA1618, TrA1622. The CD26 southern extension is altered at the base in Plan hrc1C134, and that configuration is generally the configuration found in the first public plan C125 (based on hrc1C136). US-731R; 264a PL-1138; D-512. Seliger denied instructing anyone “to keep the black population together” in CD12. TrA250. 290. On May 31, Opiela wrote to Congressman Marchant (CD24) stating, “The House and Senate have reached a compromise between their congressional redistricting maps today. The House and Senate consensus congressional proposal will likely be released today; Congressman Smith wanted me to describe your district under the proposal to you.” D-693. Marchant replied, “Eric, one change, it’s easy, no population. My grand babies go to Hockaday School on forest lane AND Inwood. I have the north side of forest, Pete has the south side. Please go across the street and pluck the campus out of Pete and put in my district. There will be no population this agreeable [sic] and I will ask Burt to do it.” Id. Opiela forwarded the request to Downton and Interiano on May 31. Id. 291. Plan C125 was released May 31. The configuration of DFW, including the CD26 lightning bolt, was drawn by Downton. TrA249 (Seliger); TrA1607 (Downton). No new Latino or coalition district was placed in DFW as had been advocated by Congress- man Smith and others. Instead, the minority population of existing minority district CD30 was increased (B+HVAP went from 76.7% to 81.5% and BVAP went from 42.5% to 46.4%) and the remainder of the urban and minority populations of Dallas and Tarrant Coun- ties were split apart and connected to suburban districts. A new district CD33 was drawn to include all of the City of Arlington in eastern Tarrant County, joined with Parker County to the west and part of Wise County to the north. 292. At the June 2 HRC hearing, Solomons laid out Plan C125 and said they had not created a Hispanic- majority district in North Texas because (1) none of 265a the proposed maps had over 50% HCVAP, whereas they were able to draw a majority-HCVAP district in Central Texas (CD35), (2) they “drew a coalition district in District 6, with a combined Hispanic voting age population 39.0 percent and a black voting age population 11.4 percent, which is exceeding 50 percent”16; (3) “we were concerned that the proposed maps did not maintain communities of interest. For example . . . Representative Veasey’s map [Plan C121] split the cities of Fort Worth, Haltom City, Arlington, Grand Prairie, Dallas, Irving, Farmers Branch, Carrollton, and Boggs Springs. Most large cities get split, but we tried to avoid splitting smaller cities whenever possible. Our North Texas districts keep Arlington, Grand Prairie, Haltom City, Farmers Branch, Carrollton, and Boggs Springs all whole. We actually tried to keep small cities and small counties whole throughout the State when we were able to.” D-601 at 10. Rep. Alonzo (Hispanic, Democrat) stated that he was “extremely disappointed” with the lack of a new Hispanic opportunity district in DFW in the proposed plan. Id. at 35. Rep. Veasey (African American, Democrat) noted that his proposed map had a DFW district with 66% HVAP. Id. at 38. He said, “The district [CD34 in DFW] that I

16 CD6 was 11.4% BVAP and 39% HVAP, and 50.1% B+HVAP. It also had an Anglo CVAP of 59.5%. D-548. Therefore, it was only a bare majority in terms of B+HVAP and was not a majority- minority CVAP district. It was inconsistent for Solomons to assert that CD6 was a coalition district based on VAP when the mapdrawers had insisted on using CVAP to determine whether a district was a VRA opportunity district. In addition, mapdrawers knew this district would not perform for minorities, and given their adherence to a 3-1 map policy, would not have drawn it to perform for minorities because they believed that would make it a Democrat district. 266a drew on the Texas Fair Plan [Plan C121] is – 100 percent follows the Voting Rights Act. I want to be as crystal clear on that as possible. It’s an effective Latino opportunity district that is 71.8 percent Hispanic with the Hispanic voting age population of 66.2 percent. And if – and when you add that with a [combined] minority citizen voting age population, it puts it at 65 percent. And I just think that that’s important to point out because I’m not sure exactly what Ryan [Downton] is looking at . . .” Id. at 128. 293. Veasey complained that all the African Americans were put into one district in Dallas County (CD30) unnecessarily. D-601 at 56-57. He further noted that although minorities accounted for the growth and provided the population for the new districts, they lost strength under the proposed map, while Anglos gained strength, and that the proposed map violated the VRA. Id. at 65. Solomons responded with his opinion that the map was an improvement for minority representation. Id. at 66. Anita Privett stated, “The [League of Women Voters] is especially concerned about the minority voters in Districts 6, 7, and 27. Under the proposed map minorities . . . comprise more than 50 percent of the population in each of these districts, yet would not have the opportunity to elect candidates of their choice based on 2008 election data.” Id. at 83. Rep. Alonzo submitted “a list of 17 people that had requested they create a North Texas opportunity district.” Id. at 133. 294. On June 2, Seliger’s statewide substitute plan C130 was made public. It made a small change to the boundary between CD12 and CD26, which Downton thought was to put an airport back into Granger’s district that had been in her benchmark district. TrA1623 (Downton). 267a 295. At the June 3 SRC hearing, Seliger said the new districts went where the population growth was, and that new CD33 was an Arlington-based district in the DFW area that included Parker County and a portion of Wise County. D-602 at 12. Sergio DeLeon from Tarrant County opposed the plan because it did not create a Hispanic opportunity district in the DFW area and it split the Hispanic community of Tarrant County into a number of districts. Id. at 48-49. With regard to DFW, Perales said, “I understand that there was some talk in the House hearing yesterday, that . . . a district could not be drawn that way because nobody had provided an . . . example of a district that was over 50 percent Hispanic citizen voting age population. I wanna make absolutely clear that such a district can be drawn. We proffered a map that was built out of whole VTDs, that was at 45 percent Hispanic CVAP, but it is absolutely possible and I know that any competent mapper who is advising the Senate Committee, can show that that district can be drawn. It is there. And, I would also like to point out that the Latino community that is encompassed in our proposed district is spread across more than three . . . districts. They are currently residing in the proposed, under [Plan] C130, District 30, District 6, District 33, District 12, and District 26.” Id. at 85. Sen. Uresti proposed Amendment 217 to Plan C130 to the DFW area, with a minority district similar to the one in Veasey’s Plan C121 and MALDEF’s Plans C122 and C123, but it did not pass. Id. at 183-84. The DFW area did not change between Plan C130 and C136, the map that the Senate passed out of committee.

17 This was likely Plan C133 or C135, which contain identical versions of the proposed minority district CD33. The Court does not have HCVAP data for these plans. 268a 296. At the June 6 Senate debate, Sen. West asked why no minority seat was created in the DFW area. Seliger stated that they tried to create a minority district in DFW and “carefully analyzed” the MALDEF suggestion, but they could not produce a performing district there and could not produce a district that met the requirements. D-22 at A-5. He stated that the shape of the district in MALDEF’s map was “odd” and would not be required by the VRA. Id. Seliger acknowledged that the Solomons-Seliger map did contain a number of districts that could be considered oddly shaped, but he thought they were not as oddly shaped because they lacked the “tentacles reaching out.” Id. at A-6. Seliger also denied that CD30, the African-American district, was packed. Id. at A-8. Seliger laid out Floor Amendment No. 1 (Plan C137), an amendment relating to precincts in Dallas County, switching a few precincts between CD32 (Sessions) and CD5 (Hensarling) in Dallas County. Id. at A-17. These changes were requested in a June 4 email from Blaine Brunson to Doug Davis and Reb Wayne and were referred to the “Sessions/Hensarling swap” in the SRC’s RedAppl account. US-625; D-533; PL-1322. The amendment was adopted. D-22 at A-17. Veasey’s Fair Texas Plan C121 and Sen. Gallegos’s Plan C131, both of which included a new minority district in DFW, were offered and tabled. Plan C141 was the Senate engrossment. 297. Solomons released his statewide substitute Plan C144 (reflected in hrc1C145 in Downton’s plan log) on the morning of June 8. It changed the bottom portion of the CD26 extension quite significantly, resulting in a configuration that was similar, but not identical to, the final configuration in Plan C185. Downton testified, “[P]eople made us aware that we had split the Hispanic community in Tarrant County, 269a that there is a north Fort Worth Hispanic community here and also a south Fort Worth Hispanic community here, and we had put them in different districts.” Tr912; Downton 8-12-11 depo. at 129-30 (noting that Plan C125 only contained the north Fort Worth His- panic community).18 Downton said that he did not use racial shading when he was initially drawing the districts, and it was after he was told that he had split the Hispanic population that he turned on Hispanic shading and saw that he had. Tr913. Downton testified that they decided to make a change to put them in the same district and essentially took out an area at the base of the lightning bolt and replaced it with another area to unite the two Hispanic areas in CD26. Tr914-15, Tr942-43, TrA1629. He stated that he used racial shading and put almost all of the concentrated Hispanic population of Tarrant County into CD26, except Hispanics in the Trinity River Project area that remained in CD12. TrA1624-26; Downton 8-12-11

18 Rep. Alonzo had an exhibit entered into the record on June 14 that complained that in Plan C141, the Tarrant County Latino north side community is exiled to Denton-based suburban Dis- trict 26, while the growing south side Latino community is placed in District 33, which is based in Arlington and Parker County. D- 23 at S54. Downton testified for the first time at the August 2014 trial that he had learned from a blog (gregsopinion.com), which made observations about redistricting, that he had split the His- panic community, and he thought he found the blog through a link on texasredistricting.org. TrA1628, TrA1710, TrA1783-84, TrA1803. He testified that the blog pointed out that what he thought of as two distinct populations were considered one community. TrA1624. The blog entry entered into evidence by Defendants, D-715, complains only generally that Plan C125 frac- tures the minority community throughout the DFW area and connects minority populations to heavily-Anglo populations in surrounding counties. There is a map of DFW shaded by CVAP population. It does not discuss the Hispanic population in Fort Worth specifically. 270a depo. at 132. Downton stated that the changes were made to help the Hispanic community by putting it together in one district; they would stay whole within a single district and be a solid voting block that anyone running for office in that district needed to care about, rather than being split. Downton 8-12-11 depo at 131- 32. The Tarrant County Hispanic community was thus placed together in Anglo-majority CD26, increasing the HVAP of CD26 from 23.3% in Plan C125 to 24.5% in Plan C144. Downton also included the African- American neighborhood of Como in CD26 with the Hispanic community because Veasey had asked for it to be in Burnam’s Hispanic district (HD90) when they were creating the House map. TrA1625, TrA1713 (Downton). Downton had no personal knowledge of the community characteristics for the residents who live inside the lightning bolt; he only knew they were Hispanic, they were in Burnam’s House district, and they lived in Fort Worth. TrA1710, TrA1745-56, TrA1789 (Downton). He used racial shading, combined with the knowledge he had gained from the House process and the discussions about HD90 regarding the Hispanic population and the Como area, to decide which population to include. TrA1711 (Downton). 298. Downton did not think that the Hispanic population needed to be joined to comply with the VRA and did not try to combine them in Plan C125 because the district was not a minority district. However, he said that “people” wanted to do that and thought they needed to for VRA compliance in that district. TrA1744-46. Downton testified that other people raised it as a concern and a possible legal challenge, so he changed it to address that concern and prevent a legal challenge on the basis that it would split up minority groups. TrA1746. He testified that he felt like it was a “damned-if-you-do, damned-if-you-don’t” 271a situation because if he did not fix the problem, people would complain that he broke up these communities of interest and if he did fix the problem, people would complain that he used race. TrA1802. He further stated, “We were maintaining the partisan nature of all of the Republican districts, but you could do that the way I did it originally without regard for ethnicity, or you could do it the way we ended up with, where you still maintained the partisan makeup but you kept the ethnic communities together.” TrA1802-03. 299. Hanna testified that he said that he thought it was better not to divide a Black or Hispanic neighborhood between different districts and offered advice that minority neighborhoods be kept whole, but that he “did not express an opinion as to whether Blacks and Hispanics should have been divided or not.” TrA1546. Hanna felt he was asked to look at how the boundaries of CD26 fell with respect to race, but did not offer any opinion about the extent to which those boundaries might separate Latino from African- American residents in Tarrant County. TrA1547-49 (Hanna). “I said I felt like the – both the Black and Hispanic neighborhoods should not be divided within that neighborhood so that the Black neighborhood should be complete. The Hispanic neighborhood should be complete, and the lines should not cut across sort of a Black area or a Hispanic area.” TrA1569 (Hanna). That was the only advice he remembered giving about the lightning bolt. Id. Hanna advised not splitting minority neighborhoods because keeping a neighborhood whole is a traditional redistricting criteria and because it would tend to dilute the vote of that group. TrA1574 (Hanna). Hanna stated that there may not be an obligation to prevent vote dilution when you are creating a district that has a fairly low minority population (i.e., is not a minority opportunity 272a district), but that he had concerns that if there was a persistent pattern of dividing minority neighborhoods, that would be a problem for the State. Id. 300. Downton said it also “got a little more compli- cated” because Congresswoman Granger wanted to keep as much as possible of the Trinity River Project, “so we tried to draw the line in a way that left as much of the Trinity River Project in her [district].” Tr915. Downton said that he learned through the course of the public hearings and further conversation with Geren that he had not gotten all of the Trinity River Project, and he told Geren that it would not be possible because there had to be a conduit to pick up Demo- cratic votes for CD26, but he would put as much as he could into CD12. TrA1624. He stated that the Trinity River Project area did not have very many people, but including it “does create a very strange shape.” Tr915; Downton 8-12-11 depo. at 132. 301. On June 8, Matt Leffingwell emailed Opiela with the subject “Changes to CD12.” D-622. Leffingwell wrote, “Our office that is located on 7th and Jones street in downtown Fort Worth was drawn out. There is no population there. We told Charlie Geren about this problem and he said he would work on this today but this has to be corrected.” Id. Opiela forwarded the email to Downton and Interiano and wrote, “Heads up. Charlie Geren is going to offer an amendment to Plan C144 to put downtown Fort Worth back in the Fort Worth Congressional District rather than the Denton County Congressional district [CD26]. Easy fix, you just have to keep the transit East of IH 35W and go under I 30.” D-622; PL-311 Pt. 5. 302. Later, Opiela emailed Interiano and Downton to inform them that Congressman Marchant (CD24) would oppose Rep. Madden’s proposed amendment 273a Plan C147 to Plan C144. PL-311. Plan C147 proposed changes to CD24 and CD26 in Denton County, and to CD3, CD4, and CD32 in Collin County north of DFW. Opiela later wrote, “Just heard from Cong. Sessions’s staff that he spoke with Rep. Madden and is now fine with C147. Cong. Johnson is as well. Cong. Marchant remains opposed.” Id. Downton replied, “Is there an amendment to the amendment that we can do to get Marchant on board?” Id. Downton agreed to an amendment that would not affect Collin County/ Marchant’s district. Id. 303. At the June 9 HRC formal meeting, Solomons laid out Plan C144 and it was adopted. Plans C121 (Rep. Veasey), C126 (Rep. Alvarado), C142 (Rep. Alonzo), which created additional minority opportunity in DFW, were offered and tabled. Rep. Madden offered Plan C147 as amended by C148, and it passed, slightly modifying CD24 and CD26 in Denton County. 304. On June 14 during the House debate, Rep. Geren (Anglo, Republican) proposed Amendment 2 (Plan C169), which he said was “a swap between Con- gressional Districts 12 and 26” to make downtown Fort Worth whole, put the entire city of Westlake in the same district, and “unite the black communities in Fort Worth.” Downton worked with Geren on this amendment. TrA1629 (Downton). The amendment took the Como area (a primarily black area) out of CD26, separated it from the Hispanic community, and put it with the black community in CD12. Id.; D-566. Downton testified that this was a request from Rep. Veasey, who thought those communities belonged together. TrA1629. Downton also testified that the amendment may have carved out more of the Trinity River Project and may have moved Congressman Burgess’s and Congresswoman Granger’s district 274a offices back into their districts. TrA1629-30. The amendment was adopted, and this was the final configuration of CD12 and CD26. TrA1630; D-23 at S3. 305. Rep. Johnson (African American, Democrat) offered Plan C157 (Amendment 8) affecting CD5, CD30, and CD32 in Dallas County, and Plan C177 (Amend- ment 9) as an amendment to the amendment, which was initially adopted, but later the vote was reconsidered and the amendment withdrawn. PL-281. Amend- ment 8 as amended by Amendment 14 (C178) (affecting CD30 and CD32) was later adopted, affecting CD30 and CD32. These changes did not increase the minority population of CD30. Compare C149 (D-559.2) with C178 (D-568.2). Plans C121 and C142 were again offered and tabled, as was Plan C155. 306. Martinez Fischer offered Plan C163, which he said had “two minority opportunities” in DFW. D-23 at S24. Solomons said, “I will point out that Mr. Martinez Fischer’s Plan 163 splits a number of cities. All maps sort of do, but in this case, in Dallas and Tarrant County alone, it splits Garland, much for the committee and me.” Id. at S23. Solomons moved to table. Martinez Fischer said that the committee map cut cities like Austin and San Antonio, so they should be consistent if that is the standard. Id. at S24. Martinez Fischer introduced Plan C164, which he argued better acknowledged minority growth. Id. at S26. Solomons opposed, stating, “It still splits a number of cities in North Texas.” Id. Solomons moved to table and the amendment was tabled. Id. at S28; PL-218 at 8. C165 was also introduced and tabled. 307. Rep. Dukes (African American, Democrat) offered Plan C166 (Amendment 17). She said it was a map that “incorporates the new Dallas-Fort Worth minority district, which is labeled CD35 in this map, 275a that is very similar to the CD34 Veasey plan, C121, and has the same Hispanic and black performance. The Hispanic citizen voting age in this district is 45.6 percent. The black citizen voting age is equal to 18.3 percent [it was actually 16.7%] with a combined percentage way over the 50 percent marker that Repre- sentative Solomons has been discussing.” Solomons opposed, stating, “I have some legal concerns about the map. It creates one less Hispanic district – Hispanic majority district – than the committee map, and neither the new District 35 in North Texas nor District 36 in Harris County are Hispanic minority districts.” D-23 at S36. At the 2011 trial, Dr. Ansolabahere noted that proposed CD35 connects two big Hispanic populations in the center of Fort Worth and the center of Dallas. Tr1140. He noted that it was well over a majority HVAP and given the population projections, it was “almost surely over 50%” HCVAP by extrapola- tion. But he would want to wait for the 2010 ACS data to answer the question for sure Tr1140-43. 2008-2012 ACS data shows 47.4 (+/- 1)% HCVAP. 308. On June 15 during third reading in the House, Plan C185 was offered and adopted. Rep. Zedler (Anglo, Republican) offered Plan C186 to make changes to CD6, CD24, and CD33. He said, “6 is Joe Barton’s. What it does is it takes him—he used to have almost all of Arlington, the current map gives him none, or very little, and this one gives him a portion of Arling- ton, including his district office, as well as his home. In the current map, his home and his district office were taken out, and I think this is a good, fair way to get this situation solved.” D-604 at 415. Solomons opposed the amendment, stating, “It’s a last minute amendment. We’re not exactly sure of all its ramifications, but we can tell at this point, it could have been filed yesterday. We’d have had time to analyze it. This 276a amendment, basically, as far as we can tell right now, splits Arlington, takes about 57,000 people out of Arlington. We made a concerted effort to try to keep Arlington whole and to keep the cities whole when possible. It does more than just the district office and an alleged house. Right now our records show that Mr. Barton -and this is all about Mr. Barton, to the detriment of Kenny Marchant and Congressional Dis- trict 24 - he literally takes Hurst out of Congressional District 24 for Congressman Marchant, who was a former member here. He wants to put in Ranger Sta- dium - he already has Texas Stadium - apparently he needs the ballpark stadium, as well. And, quite frankly, he moves Hurst out of Congressional District 24 to help get his population, as well. It seems to me that if Mr. Barton had wanted to do this earlier and work with us through the second reading amendments or even before, it would have been more appropriate. I do think it has some other ramifications. I would ask you to table this amendment, and we’ll send this bill over to the senate and see what else we need to do with the map, if necessary. But I do, dramatically, think that this is the epitome of Joe Barton, Congressman Barton, wanting to have just exactly what he wants without really going through the process, as all of the other congressmen and everybody else really did. This last-minute amendment does dramatically affect some other districts.” Id. at 418. It was tabled. Downton testified that there were items that Barton wanted that he did not get, including the Arlington ballpark, and that certain areas were taken from him to create the new CD33 based in Arlington. TrA1698-99. 309. Rep. Alonzo asked which congresspersons had input into the map, and Solomons said that both Republican and Democrat congresspersons had input. D-604 at 419. When asked about a new opportunity 277a district in DFW, Solomons said there was “a difference of opinion” and “a lot of it had to do with the assimilation of Hispanics and blacks throughout the community in North Texas, and the Hispanic citizen voting age population issues. So when you start looking at those numbers, they become kind of important.” Id. at 419. Alonzo included a petition for a Latino opportunity congressional seat in the DFW area, anchored in Dallas County, in the record. Id. at 421-24. 310. The CD26 lightning bolt splits 38 VTDs in Tarrant County. TrA409 (Arrington); PL-1635; PL- 1633. Downton testified that some splits were required but others were not. TrA1631. He testified that many of the splits (including Precinct 4138) were due to equalizing the population because he would go around the borders of the districts and look for blocks the size he needed until he got to zero. He said that every time he changed the map, he would do it again, such that were a lot of splits in areas that he worked on more regularly, like Tarrant County. TrA1718. He also testified that the changes for district offices and to move an airport at the block level would have split precincts. TrA1631-32. He said that the splits in Precinct 4015 were to put the three census blocks that touched the Trinity River into CD12. TrA1719; PL-1154. He denied splitting precincts in DFW with discriminatory intent. TrA1632. 311. The CD26 lightning bolt from Denton County into Tarrant County picks up areas that are disproportionately Latino and puts them in a Denton County district that is overwhelmingly non-Latino so they will have no opportunity to elect. Tr257 (Kousser). The lightning bolt of CD26 intentionally joins a significant Latino population in Fort Worth with another significant Latino population area to the south. TrA1712, 278a TrA1718 (Downton); PL-1145; PL-1150. Part, but not all, of the strange shape in the area adjoining the two Hispanic populations is explained by the goal of including as much of the Trinity River Project in CD12 as possible. TrA1198 (Moss); D-682; D-683; D-684; D- 711; Quesada-115. Some Hispanics are not included in the CD26 lightning bolt due to constraints imposed by the Trinity River Project but otherwise the boundaries of the lightning bolt closely track areas with majority- HCVAP areas. Quesada-379 & 388. There are split precincts where mapdrawers split off the Hispanic population and put it into CD26, so the lines were clearly drawn to a large extent along racial lines. TrA410, TrA472 (Arrington). Downton was purposefully trying to keep the Fort Worth Hispanic “community of interest” together even though CD26 was not a Latino opportunity district. TrA1720 (Downton). 312. CD12 splits 53 precincts in Tarrant County. The African-American community of Como was drawn out of CD26 and into CD12. TrA1714 (Downton); PL-1146. In that area, the CD12/CD26 boundary also splits Precinct 1120 in a way that includes more Hispanics inside CD26 than out and excludes more African Americans than are included in CD26, and Downton could not explain the split other than to speculate that it was zeroing out or a district office. TrA1715-16 (Downton); PL-1151; PL-1152. African- American populations are generally excluded from the CD26 lightning bolt, including by split precincts. TrA1714-17 (Downton); PL-1146; PL-1147; PL-1148; PL-1149; US-643_0003. The boundaries of the CD26 lightning bolt (particularly the southern portion) are more consistent with race than with political affilia- tion, with majority-HCVAP areas inside CD26 and majority-BCVAP areas outside CD26 (and in CD12). D-657; D-640; D-641; D-702; Quesada-25; Quesada-27; 279a Quesada-32-35; Quesada-340; Quesada-379; Quesada- 388; US-642; US-643; US-722; PL-1630. 313. CD6 splits 39 precincts—23 in Dallas County and 16 in Tarrant County. In Dallas County, there are precincts splits (1105 and 1127) where more Hispanic population is inside CD6 and less Hispanic population is outside CD6. TrA1721 (Downton); PL-1155. Downton stated that the splits were caused by zeroing out because there were several in that area and he was not trying to put Hispanics into CD6 since it was not a minority district (and nearby Hispanic areas are not included). TrA1720-21. Similarly, in Precincts 4615 and 4614 more Hispanic population is in the district and less Hispanic population is outside. TrA1722 (Downton); PL-1156. Downton testified that these were also caused by zeroing out because he would not have drawn jagged one-block lines and there were a series of blocks pulled out. TrA1722. Where this supposed zeroing out occurred, it still tended to move relatively lower Hispanic population blocks (as a percentage) out of CD6 in 4610, 4615, 4614, and 4433 and keep higher Hispanic population areas in the district. TrA1722-24 (Downton); PL-1156; PL-1157. Further, zeroing out does not account for the high number of precinct splits, especially given that Dallas County was not drastically changed from Plan C125 to Plan C185. 314. Plan C185 packs CD30, the only minority district, by increasing its minority population even though it was already consistently performing and there was no indication that increased minority population was needed to maintain its ability to elect. TrA1397 (Murray); Tr1276, 1281 (Johnson). From the benchmark, BVAP increased from 42.5% to 46.5% and HVAP increased from 34.7% to 35.6% (B+H total population 280a went from 81.3% to 85.2%). BCVAP increased from 49.8% to 53.1%. CD30 splits 31 precincts in Dallas County. CD30 retained 80.7% of its benchmark population. D-403. 315. Plan C185 divides (“cracks”) the remaining urban and suburban minority population in Dallas and Tarrant Counties through bizarrely shaped fingers and puts them into Anglo-dominated rural and suburban districts where their influence is minimized. Tr672 (Korbel); Tr257 (Kousser); TrA407-08 (Arring- ton); TrA1397 (Murray); Joint Expert Ex. E-2 (Kousser report) at 119-27; Joint Expert Ex. E-11 (Korbel report) at 4. There are four essentially suburban districts that move into Dallas County and pick up about 624,000 Hispanics from Dallas County. TrA408; US- 352 (Arrington report) Table 16. The plan strands urban Hispanic and African-American voters in Anglo- dominated suburban/rural districts. Tr1130 (Ansola- bahere). The DFW districts link inner-city neighborhoods with rural areas, protecting Anglo Republicans at the expense of minority voters. Tr1044 (Murray). The Anglo voters have a history of polarized voting against minorities and high turnout that will over- whelm the minority voters in these inner-city neighborhoods. Id. For the life of the plan, minority voters will not be able to extend their influence outside CD30. Id. 316. New CD33 includes all of Wise County, part of Parker County, and then takes a portion of the minority community in the southern part of Tarrant County and reaches around and picks up all the minority population in the center city area of Arling- ton. Tr663 (Korbel); Joint Expert Ex. E-11 (Korbel report) at 3; Quesada-46. The Tarrant County arm includes approximately 300,000 persons that are 29% 281a Anglo and 64% Hispanic and African American. Tr664 (Korbel); LULAC-16A. 317. Similarly, as discussed, there is a concentrated minority population in the middle and southern portion of the CD26 “lightning bolt” extending into Tarrant County from Denton County. That portion of the lightning bolt contains 134,829 persons that are 76.5% Hispanic and African American, and is tied to a heavily Anglo area. Tr665 (Korbel); LULAC-16B. The Tarrant County population from CD12, CD26, and CD33 is 693,907 persons, of which two-thirds are minority. Tr668 (Korbel). 318. CD6 includes two predominantly Anglo suburban counties (Ellis and Navarro) and then splits to the west into the balance of the minority population in Tarrant County and to the east almost to the center of the city of Dallas in Dallas County, where it picks up substantial minority population. Tr666 (Korbel); Tr1130 (Ansolabahere); LULAC-16F; Joint Expert Ex. E-15 (Ansolabahere report) at 12; Quesada-399. CD5 similarly joins Anglo rural counties with substantial urban minority population in eastern Dallas County, and the communities have very little in common. TrA690 (Johnson); LULAC-16H. 319. In Plan C185, CD30 has 16.6% Anglo VAP, while CD5, CD12, CD24, CD26, CD32, and CD33 are all around 60% Anglo VAP (CD6 is 44.4% Anglo VAP but 57.7% Anglo CVAP). 320. Dr. Ansolabahere notes that Hispanics are 38% of the Dallas County population, Whites are 33%, and Blacks are 22%. Of the five Dallas districts (5, 6, 24, 30, and 32), none are majority or plurality His- panic, and thus none of the Hispanics in Dallas County are in a majority or plurality-Hispanic district. Blacks 282a are a plurality in one district (CD30), which is about equal with their population share. Even still, 58% of Blacks are in CD30 and 42% are in Anglo-majority districts. Joint Ex. E-15 (Ansolabahere report) at 22. Whites are an outright majority in 3/5 districts (5, 24, and 32) and the plurality of the VAP in CD6 (which is majority Anglo CVAP). Over 90% of Whites live in districts in which Whites are the majority. Id. Whites are 52% of Tarrant County, Hispanics are 27%, and Blacks are 14%. Id. at 24. Tarrant County residents are the majority of two districts (12 and 33), and these are Anglo-majority. Parts of Tarrant County are in four additional districts (6, 24, 25, and 26), and those are all majority White. None of the 270,000 Blacks or 480,000 Hispanics in Tarrant County are in a district where they are a plurality or majority. Id. at 24. 321. Demonstration Plan C188 creates two coalition districts in DFW in addition to maintaining CD30 as a 47% combined BCVAP district. CD12 is 38.4% BCVAP and 20.6% HCVAP and CD35 is 41.1% HCVAP and 17.7% combined BCVAP using 2008-2012 ACS data. Joint Map. Ex. J-10. 322. Demonstration Plan C192 creates CD34 and CD35 as new minority districts. CD34 is 45.6% HCVAP and 16.7% combined BCVAP using 2005-2009 ACS data; CD35 is 15.2% HCVAP and 35.6% combined BCVAP; CD30 is 45.6% combined BCVAP. Joint Map Ex. J-13. 323. The NAACP submitted demonstration Plan C193 (a 7-district plan), which creates three minority districts in the DFW area. Joint Map Ex. J-14. CD30 is maintained as an African-American district with 50% combined BCVAP using 2005-2009 ACS data. CD34 would be 15.8% HCVAP, 32.4% Black alone CVAP, and 46% White alone CVAP. D-571.3. CD35 283a would be 44.6% HCVAP, 15% Black alone CVAP, and 37% White alone CVAP. Id. Neither one has a single majority minority population. TrA1194 (Moss).19 Moss testified that proposed CD34 joins African-American communities with commonalities and provides an opportunity to elect an African-American candidate. TrA1885. Moss stated that the proposal was preferable to the current Court plan because it has a His- panic and an African-American district. TrA1187. Congresswoman Eddie Bernice Johnson testified that CD34 would be an effective African-American district and CD35 an effective Latino district. Tr1292. She testified that African Americans and Latinos work in coalition and it would be appropriate to pair them in CD35, and that African Americans and Latinos would work with Asians in CD34. Tr1293-94. Murray testified that CD34 and CD35 would be effective opportunity districts, with CD34 being an effective district for African American voters, and CD35 being an effective district for Latinos. Tr1035-36. Fairfax estimated CVAP for CD34 and CD35 for 2010 as follows: CD34 is 17.2% HCVAP and 34.6% BCVAP (51.8% H+BCVAP); CD35 is 46.7% HCVAP and 16% BCVAP (62.7% H+BCVAP). TrA801. Table 4 gives the 2008- 2012 ACS CVAP estimates as: CD34 is 17.7% HCVAP and 35.2% BCVAP (52.9% H+BCVAP); CD35 is 45.8% HCVAP and 15.8% BCVAP (61.6% H+BCVAP). Fair- fax also opined that CD34 and CD35 were compact enough to avoid Shaw problems. TrA795.

19 The NAACP offers Fairfax’s testimony that CD35 would have been 51.92% HCVAP in 2014. TrA805 (NAACP fact finding 191). In 2010, the estimated HCVAP was 45.8 (+/- .9)%. D-571 (2008-2012 ACS data). Fairfax noted that CD35 would have become majority HCVAP sometime between 2012 and 2013. TrA806. 284a 324. African-American lay witness Franklin Douglas Moss, Sr. from Tarrant County is familiar with voting in the area because he lives there and has worked on political campaigns. TrA1175 (Moss). He testified that it is very difficult for an African-American or Hispanic candidate to win unless they have a coalition of minority support. Id. He cited examples of coalition voting, including school board elections, city council elections, and Wendy Davis for state senate. TrA1176. He testified that Mark Veasey was elected in the new DFW district in the interim plan with the support of the coalition, and he could not have won without Hispanic support. TrA1179. Moss testified that Black and Latino voters share common concerns and interests, ranging from economic development issues, housing issues, health and health disparities, employment issues, housing, etc. TrA1180. 325. Moss testified that C185 splinters the minority communities in Tarrant County, specifically those that normally have worked together in supporting candidates. CD12 splinters the African American and His- panic communities and CD26 comes in and pulls the Hispanic neighborhood out of Fort Worth. TrA1181 (Moss). Moss testified that combining urban minority communities with suburban or rural areas was problematic because the voters do not have a lot in common economically and do not have the same concerns, and the rural and suburban populations will dominate. TrA1181-82. These concerns are present in CD33, CD26, and CD6. TrA1183. 326. Moss testified that CD33 in the Court’s interim plan (C235) gave minorities an opportunity to elect someone from Fort Worth that looks like the people in the district. TrA1184. Moss testified that the previous representative of his neighborhood, an Anglo (Burgess 285a (CD26)), did not live in the area and Moss’s African- American neighborhood was just a small part of his district and did not determine whether he would be reelected. TrA1179. Moss testified that Veasey is involved and responsive. Id. Moss stated that Con- gresswoman Johnson of CD30 in Dallas had been the only representative that they could go to, and he had been referred to Congresswoman Johnson on issues by Tarrant County congresspersons. TrA1185-86. 327. Moss also testified that there has been voter intimidation at the voting precincts where there have been critical races and minority candidates running against Anglo candidates. TrA1177. He said that the Anglo candidates sent various poll watchers and other people to polling locations in predominantly black precincts to intimidate individuals and tell them that they are doing things illegally when they are not. Id. Moss was unable to name specific elections, but said it had “been a whole series of elections,” including recent elections. Id. Moss learned about it from discussions at meetings of the coalition of black Democrats, which precinct chairs attend. TrA1178. The hearsay objection to this testimony is sustained to the extent it is offered for its truth. 328. Murray testified that as the Hispanic and African-American populations in DFW increase, they are having a much higher success rate in winning offices in Dallas County. TrA1401. 329. In 2012, there was a runoff election between Domingo Garcia, a Hispanic from Dallas County, and Marc Veasey, an African-American from Tarrant County, for the new CD33. TrA1401 (Murray). Murray found that both race and the “home factor” affected voting in the runoff. Id. Although each minority group tended to vote for the candidate who looked like them, 286a both groups supported Veasey in the general election. TrA1402 (Murray). Murray thought it preferable to draw districts that separated the African American and Hispanic candidacies. Id. 330. Despite the fact that minority growth accounted for the new congressional seat, the new district (CD33) was placed in a predominantly Anglo area. No new minority districts were created in the DFW area. Although numerous proposals, including from the Republican congressional delegation and the Gover- nor, included a minority district in DFW, mapdrawers rejected such a district because it was not 50% HCVAP and they regarded it as a Democrat district. No map was introduced during the session that had a 50% HCVAP using 2005-2009 ACS data, but maps with one or two new coalition districts (exceeding 50% B+HCVAP) were submitted. Because any coalition district was regarded as a Democrat district and mapdrawers had resolved to create a 3-1 map, it was decided early in the process that no new minority district would be drawn in DFW, even though it would make the Republican districts in the area stronger. 331. The minority population of CD30 was increased despite the fact that it reliably performed and there is no evidence that an increase in minority population was necessary to maintain its ability to elect. The evidence indicates that CD30 was intentionally packed with minority voters. 332. The remaining minority population in urban Dallas and Fort Worth was cracked by various fingers and extensions from Anglo-majority rural and suburban districts reaching into the urban areas and stranding the minority population in majority-Anglo- CVAP districts. Mapdrawers intentionally packed and 287a cracked minority populations in DFW. The progression of plans in Downton’s plan log shows that he intentionally split the proposed minority district among the various Anglo-dominated districts in DFW, and that changes were later made in Tarrant County to accommodate Congresswoman Granger’s request to include North Richland Hills and the Trinity River Project in her district and to unite two Hispanic communities in Fort Worth. 333. Minorities are not represented in proportion to their numbers in the DFW area. 334. Downton admitted to using racial shading and placing certain voters into CD12 and CD26 based on their race to keep racial communities together, even though CD12 and CD26 were not minority-ability districts and he did not need to do so to comply with the VRA. Congress - CD27/Nueces County 335. From 1968 to 1980, Nueces County was joined with other coastal counties going north. In 1982, it was first joined in a district going south and including Cameron County. TrA1112 (Hunter); LULAC-3. Solomon P. Ortiz (Hispanic, Democrat) was first elected to CD27 in 1982 and began serving in 1983. TrA1113 (Hunter). Nueces County remained in the southern configuration until the 2011 redistricting. 336. Benchmark CD27 began in the eastern part of San Patricio County and moved south along the coast to the Valley, including all of Nueces, Kleberg, Kenedy, and Willacy Counties, and part of Cameron County. Downton 8-31-11 depo. at 50. The Cameron County population in CD27 was 309,545 and the Nueces County population was 340,223. TrA1077 (Hunter); D-400.2. Benchmark CD27 was a majority- 288a Hispanic district and was a performing Latino opportunity district. Tr970 (Downton); Tr514 (Engstrom); Engstrom Corr. Rebuttal Report (docket no. 307-1) at 25. 337. At the Corpus Christi field hearing on July 21, 2010, a number of witnesses testified about congressional redistricting. TrJ635 (Herrero); TrA1071 (Hunter); D-574 (transcript, with incorrect date of June 21, 2010); D-116 at 33; US-272A; US-460 (same). There were two co-chairs, Rep. Herrero and Rep. Hunter. TrA1072 (Hunter). Nueces County Judge Lloyd Neal testified about the community of bay-area counties, specifically San Patricio County, Nueces County, and Aransas County. D-574 at 19. Aransas County Judge Burt Mills said, “We think our little community, Nueces County, San Patricio County, Calhoun County and . . . Refugio County, we all work together well and we look forward to working together.” Id. at 22-23. Hugo Berlanga testified that a 70-mile radius around Corpus was a community of interest that shared a media market, but there were problems going beyond that, forcing candidates to contend with multiple media markets. He testified that the district should only have one port because the Corpus (in Nueces County) and Brownsville ports (in Cameron County) were competing for funds. Id. at 24- 25. Suzy Luna-Saldana asked that Latino s of interest be maintained and that the Hispanic vote not be diluted. Id. at 47-53. Foster Edwards of the Corpus Christi Chamber of Commerce testified that the Chamber thought the community of interest was “the ten or 12-county area around Nueces County . . . as far away maybe as Bee County up to maybe Victoria County down as far as Kleberg County . . . .” He said, “[D]own south there’s an area there that is agriculture and – and port-related and – and we have a lot – we 289a worry about windstorm insurance, hurricanes, and we have a lot of the same demographics. I mean, all of that kind of rolls together. It’s economic as much as anything. The job creation and the nature of the jobs and the nature of the community is –sort of together. We want to keep that area together.” Id. at 55-56. Alicia Bookout testified, “I want to see our Coastal Bend region properly represented. Having lived in the Rio Grande Valley for many years, I know that the two areas don’t share many common interests.” Id. at 57. Attorney Richard Bianchi of Aransas County testified, “[O]ur interests are certainly with Nueces and San Pat[ricio] and west and northward as those issues of water quality and transportation. Those – those are the areas that directly impact into our community.” Id. at 83. 338. Blake Farenthold, then a candidate for CD27, testified that benchmark CD27 “does a disservice to both Corpus Christi and to Brownsville. It’s very difficult for one Representative to serve two masters.” D-574 at 90. He stated that Corpus Christi and Brownsville were competitors. Id. at 91. He said, “Brownsville has more in common in a community of interest with Harlingen [also in Cameron County but not in CD27], which is right up the street, but it’s not in the district. And Corpus Christi has a whole lot more in common with San Patricio County, all of which is not within the district.” Id. He also said, “You know, we don’t read the same newspapers as they do in Brownsville. We don’t watch the same television stations. We don’t listen to the same radio stations. While San Patricio County, Aransas County, Bee County, Kleberg County, we all read the Caller Times up here, and we all listen to the same radio stations, we all watch the same T.V. stations. And as far as the Voting Rights Act, our general area would fit perfectly. 290a We’ve got an appropriate balance of majority and minority population that would probably create a minority/majority district.” Id. at 92. Rose Meza- Harrison, a Nueces County Democratic Party Chair, noted that some had testified about economic concerns and agreed that “our community of interest is based around the media, like – like a lot of people were talking about. In this – in Nueces County we call it the big three area, you hear news all around this Coastal Bend area from Aransas County, Nueces County, Kleberg County, Duval County [this County is to the west and not in the coastal bend]. That’s the community of interest that I believe needs to make up the representation for Nueces County and this area. And I don’t want to see it split.” Id. at 98. She also testified that Hispanic congressional districts should be maintained. Id. at 97. Joel Youeld testified, “Nueces, San Patricio, Kleberg and Aransas Counties are the heart of this concept. Together this population would be around 450,000 or something like that. The following counties could be included Bee, DeWitt, Goliad, Calhoun, Live Oak, Karnes, Jackson, Refugio and Victoria [all counties to the north]. This would bring the district population up to about the 675,000 level . . . .” Id. at 109. Hughie Fischer said that Aransas County should be with Nueces County, not with the Houston or Galveston areas. Id. at 126-28. Arthur Grando testified that Nueces did not have a community of interest with the Valley, but did with Aransas, Bee, Calhoun, Goliad, and Victoria Counties. Id. at 136-37. 339. Brian Bode testified, “I . . . stand in opposition to what pretty much everyone else here today has stood for, and that’s a district that’s Corpus Christi- centered with all the population areas around it, because simply put, there’s not enough population in 291a this area to support such a district. . . . Even with the framework that Mr. Youeld established earlier we would still be 25,000 people short. And, therefore, we would have to go down south, include the King Ranch and go down to Harlingen to pick up that population. So no matter what, even if this framework is established that everyone here has been advocating for, we will still have to have citizens in the Valley be in this Congressional District. Otherwise, we have to go up north and pick up rural districts including Wharton County and farther and farther north.” D-574 at 129- 30. He also noted that the proposed configuration would make it impossible for Hispanics to elect their candidate of choice. Id. at 131. 340. Rep. Hunter had heard sentiments about having separate congressional districts for Nueces/Corpus Christi and the Valley before the interim hearing. TrA1076 (Hunter). He testified that there had been discussions through the years that the Valley should have their congressional seats and that Nueces/ Corpus Christi should have their own congressional seat because they have two different ports, different trading areas, different travel, tourism zone, and different media markets, and one is primarily a coastal zone with coastal issues, a little bit different than the other. TrA1072-73, TrA1076. Hunter stated that such discussions included Rep. Oliveira and Sen. Lucio, in particular. TrA1077. The port is a major economic engine in Corpus Christi. TrA1077 (Hunter) 341. In the 2010 general election, Farenthold, an Anglo Republican, was elected to represent CD27, narrowly defeating 27-year incumbent Ortiz. Tr973 (Downton). Farenthold’s margin of victory in 2010 was very close. Id. Farenthold received 47.84% of the vote 292a (51,001 votes) and Ortiz received 47.11% of the vote (50,226 votes). D-35. 342. Dr. Ansolabahere found very high Anglo voter cohesion in Nueces County. TrA943. Anglo support for minority-preferred candidates is very low; only about 10 to 15% of Anglo voters support minority-preferred candidates. TrA943 (Ansolabahere). Dr. Engstrom found racially polarized voting in the 2010 general election for CD27. Tr510; Engstrom Corr. Rebuttal Report (docket no. 307-1) at 25. Ortiz received an estimated 86.6% of the votes cast by Latinos and only 15.9% of those case by non-Latinos. The Latino voter turnout was 46.72%. Tr510 (Engstrom) (Latinos were 46.72% of the people turning out to vote); Engstrom Corr. Rebuttal Report (docket no. 307-1) at 25-26. Ortiz was the Latino candidate of choice, and Farenthold was not. Tr226 (Kousser); Joint Expert Ex. E-2 (Kousser report) at 47; Tr515 (Engstrom). Farenthold was not elected with heavy Hispanic cross-over voting, and the race in fact demonstrated higher rates of racial polarization. Joint Expert Ex. E-15 (Ansolabahere) at 35. 343. Mapdrawers knew that Farenthold would have difficulty being re-elected in the benchmark CD27 configuration. Tr1462 (Interiano); Downton 8-31-11 depo. at 52-53. On November 20, Opiela forwarded an email he wrote to Lamar Smith and Interiano to Denise Davis, Straus’s Chief of Staff, which said in part, “Gerardo [Interiano] and I were talking last night about the problems inherent in trying to protect both Farenthold and Canseco. It will be INCREDIBLY difficult to not have a packing claim and enhance Farenthold’s reelectibility. There is a ripple effect created by splitting Nueces from Cameron County in that the Cameron County district will have to go North to pick up Anglo voters if it doesn’t pick them up in 293a Nueces, and there simply aren’t enough Anglo voters outside of Nueces in South Texas to pick up without reaching up to Wilson/Guadalupe/Eastern Bexar County.” US-76. Mapdrawers felt that splitting Nueces County from the district would mean that the Cam- eron County-based district that was left would have to stretch far north to avoid being “packed” with Hispanic voters (and that is what happened). 344. January 9, 2011 Opiela forwarded Interiano and Doug Davis an article saying Cameron County legislators wanted a new Valley seat; Oliveira is quoted as wanting a Cameron County-based district, and saying that Cameron County did not need to be divided between two congressional seats (as it was in the benchmark). D-607. 345. Downton was responsible for the drawing of Nueces County and CD27 in the congressional plan. TrA1632 (Downton). Benchmark CD27 was overpopulated by 43,500 people, and could have just shed population with few changes, but instead was almost completely reconfigured. Tr458 (Flores). Downton characterized the goals as trying to create a Republican- leaning district anchored in Nueces County (where Farenthold lives) (to increase Farenthold’s chances of being reelected) and to create a new South Texas minority opportunity district to replace CD27, anchored in Cameron County. Tr971, Tr1022, TrA1632-33; Downton 8-31-11 depo. at 49 (they “were trying to draw [Farenthold] into a Republican district”). He felt these goals were met. TrA1634. 346. From the beginning of the process, Downton was drawing a district with Nueces County going to the north, with Solomons’ approval. TrA1773 (Downton). He said he got the concept to make Nueces County the anchor of a district and give Cameron County its own 294a district early on (before the census data was available) from conversations with Rep. Hunter about the interim hearings, from looking at summaries of the hearings, and from statements from representatives and senators from Cameron County asking for their own district. Tr1022, TrA1633-34, TrA1761-62, TrA1725. By anchor, he meant the largest county in the district. TrA1762. Downton said the senators and representatives who wanted a Cameron-anchored district were Democrats Sen. Lucio, Rep. Lucio, and Rep. Oliveira. Tr1022. CD27 was drawn to include all of Nueces County, half of adjoining San Patricio County to the northwest, as well as coastal counties Aransas, Refugio, Calhoun, Jackson, Matagorda, and also Victoria, Wharton, Lavaca farther inland, and parts of Gonzales, Caldwell, and Bastrop Counties in central Texas. It was no longer a Latino opportunity district. Tr971 (Downton); Tr458 (Flores). The old and new districts overlap in Nueces County and part of San Patricio County but otherwise are completely different. Tr971 (Downton). 347. There are 340,223 persons in Nueces County, including 206,000 Hispanics, and half of the Nueces County registered voters are SSVR. TrA1109 (Hunter); LULAC-3. Nueces County Hispanics have elected their representative of choice in CD27 since 1982. Under the 2011 plan, Nueces County is added to a group of twelve heavily Anglo Central Texas counties. Nueces County is majority-HCVAP but none of the other counties in the district are. D-181. The 206,000 Hispanic residents, 195,000 Hispanic citizens, and 133,370 Hispanic citizens of voting age of Nueces County that used to be in a Latino opportunity district are now in a majority-Anglo-CVAP district in which they cannot elect their candidate of choice. The new CD27 is 51.4% Anglo CVAP, 36.8/36.7% SSVR, and 295a performs 0/10 on the OAG10. CD27 is an example of where Hispanics (approximately 200,000) are ending up “stranded” in white majority districts in Plan C185. Tr1131-32 (Ansolabahere); Joint Expert Ex. E-15 (Ansolabahere report) at 27; TrA1403, TrA1427 (Murray). 348. New CD34 was drawn to include all of Cameron County and part of Hidalgo County to the west (Hidalgo County was split among three districts), and it stretched north all the way up to include half of Gonzales County, taking all of Dewitt, Goliad, Bee, Jim Wells, Kleberg, Kenedy, and Willacy Counties and part of San Patricio County. For VRA purposes, CD34 was drawn to replace CD27 as a Latino opportunity district. TrA229-30 (Seliger); Tr971, TrA1773 (Downton). The new CD34 is 71.7% HCVAP and 71.7/71.9 SSVR, and performs 10/10 on the OAG10. 349. Rep. Hunter testified that San Patricio County, Aransas County, Bee County, and Kleberg County are coastal counties (although Bee County is not on the coast). TrA1096 (Hunter). Although Hunter and a number of witnesses noted a community of interest between Nueces and Kleberg Counties, Plan C185 puts them in different districts. Further, a number of witnesses identified a community of interest between Nueces County and San Patricio County, but San Patricio County is split, with much of it being put into CD34. Further, CD27 goes northeast and north to pick up a number of additional counties, such as Bastrop, Caldwell, Gonzales, Wharton, Matagorda, and Lavaca Counties, that no one suggested should be included in the same district with Nueces County/Corpus Christi. TrA1097 (Hunter). 350. Interiano testified that Nueces County was put in CD27 to be the anchor of the district, separate from 296a the southern areas, so that Nueces/Corpus Christi could control the district. Tr1462; Interiano depo. at 113-115. However, Interiano did no analysis to see whether Nueces County could control the election in the new CD27 and did not know if it could. Tr1462 (Interiano); Interiano depo. at 114. Mapdrawers did not analyze whether severing Nueces would reduce Nueces County’s proportion of registered voters in the new CD27. Tr1461 (Interiano). In benchmark CD27, Nueces County’s registered voters made up just over 50% of the total registered voters. Tr971-72 (Downton); Downton 8-31-11 depo. at 50-51. In new CD27, Nueces County voters are no longer the majority of registered voters in the district. Tr972 (Downton); Downton 8-31- 11 depo. at 54-55. 351. Hunter noted Nueces County is approximately four times bigger than the other counties within the new CD27, and he thought Nueces County would have a say in who is elected to CD27 in C185. TrA1078. That is something Hunter, as a long-time Nueces County resident, has been interested in. Id. CD27 was Hunter’s primary focus area in 2011 redistricting. Id. Hunter believes the C185 configuration of CD27 was a good thing for Nueces County and was responsive to citizen input. TrA1088. Hunter communicated with the HRC, Speaker staff, Congressmen Farenthold and Smith, TLC, and the various lawyers involved and said “it would have been a general message . . . that Nueces should be in a district going north.” TrA1118-19 (Hunter). Hunter did not look at whether it would be a majority-Latino district, “because knowing the counties that were proposed, it would not be.” TrA1106 (Hunter). He agreed one could create a majority- Hispanic district including most of Nueces County if the district went south. Id. 297a 352. Seliger testified that the reconfiguration of CD27 was to help Farenthold get re-elected. TrA229. He agreed that it was possible to split Nueces County so that only the portion with Farenthold’s home would be in the new CD27 going north. TrA230. Downton also acknowledged that they could have included part of Nueces County in the new CD34 and still met the goals of drawing a Republican district for Farenthold and drawing a separate Cameron County district. Tr1023 (Downton). In Task Force Plan C190, part of Nueces County is in a district to the north for Farenthold and part is joined with a district going to the south to maintain the bulk of its Latino population in the envelope. Tr460 (Flores); Downton 8-31-11 depo. at 55.20 Downton said they considered maps with Nueces County in the South Texas configuration, which would have made it easier to reach SSVR and HCVAP goals in CD20, CD35, and CD23, but they decided to do something different. Tr949-50. He stated that, if the maps had the same number of opportunity districts (as defined by 50% HCVAP), then it was a

20 Other maps also split Nueces County between a north and south district. E.g., C126, C142, C164, C187, C188, C225 (Farenthold proposed interim remedy). (LULAC also offers proposed Plan C200, which has part of Nueces County in CD27 going south joined with Cameron County, four districts coming out of the Valley counties, and seven HCVAP-majority districts in the envelope without going into Travis County. Joint Map Ex. J-20; LULAC 12-E-2. Plan C200 is a 7-district plan. The districts are not quite at ideal population and at least one district (CD33) is not contiguous.) Splitting Nueces County and keeping part of it in the southern configuration with Cameron County allows those Latinos to remain in a Latino opportunity district and permits the neighboring district to remain more compact and not stretch so far north. 298a political choice which configuration to use. Downton 8- 31-11 depo. at 63. 353. Rep. V. Gonzales noted that, despite the large scale population growth throughout South Texas, no new Congressional district was created. V. Gonzales Decl. (docket no. 331) at 6. Lay witness Hidalgo County Judge Ramon Garcia testified that the population of the Rio Grande Valley is growing and should anchor four Congressional seats (two in Hidalgo, one in Cameron, and one in Webb County) instead of three, to get better representation. Tr630-31. Webb County is not part of the Rio Grande Valley, but he testified that they share a lot of issues. Tr635. He also complained that, instead of drawing the districts such that someone from the Valley/South Texas could get elected, Plan C185 divides Hidalgo County among three districts, giving less opportunity for someone from Hidalgo County to be elected. Tr633. 354. Dr. Flores notes that Nueces County is majority Hispanic. Tr458. Traditionally it was in a Latino majority/opportunity district, but in Plan C185 Nueces County Hispanics are in the minority in new CD27. Tr459 (Flores). Dr. Ansolabahere testified that most of the Hispanic CVAP growth occurred in “the envelope” of southern and western Texas. The “envelope” is the large triangular area contained by a line starting in Nueces County, running south to Cameron County, then along the Rio Grande River to El Paso County, then from El Paso County to Bexar County, then northeast to the Hays/Travis County line, and back to Nueces County. TrA931 (Ansolbahere). In Ansolabahere’s April 28, 2014 report (Rod-913), it includes the six benchmark congressional districts 15, 16, 20, 23, 27, and 28. In Plan C185, Nueces County was removed 299a from the “envelope.” TrA942 (Ansolabahere). Remov- ing Nueces County from the South Texas configuration of opportunity districts reduces the possibility of drawing additional Latino districts. Tr459 (Flores). 355. Although the HCVAP population in Nueces County increased from 50.97% to 55.87%, Alford thought this increase was less than 14,000 people. TrA1845. However, Defendants’ exhibits indicate that the HCVAP increase was 23,795 (from 109,575 in 2000 to 133, 370 in 2010). D-180; D-181. Alford agreed there are alternative ways you can use the Nueces County population in creating districts, but he did not see any plan that used that population to create an additional HCVAP-majority district beyond the number of districts created by the state. TrA1845. 356. CD27 was a reliably performing Latino opportunity district until 2010. After Farenthold, an Anglo Republican, was elected in 2010, mapdrawers completely redrew the district into an Anglo-majority- CVAP district to protect Farenthold. To replace CD27, they drew a “new” CD34 that stretched from Cameron County, around the base of CD27, and into central Texas. Despite the population increase in South Texas and the Valley, mapdrawers did not draw a new Valley-based district (keeping the number at three). The three Valley-based districts are long, narrow districts that each reach up into central Texas, and Hidalgo County is split among the three districts. In the benchmark, only two Valley-based districts stretched up into central Texas, and CD15 did not stretch as far as the new CD34 does. 357. Although it was not necessary to include all of Nueces County in the new CD27, and other plans were presented that split Nueces County to allow CD27 to include Farenthold’s home and put the rest in a 300a southern district, mapdrawers made the “political decision” to put all of Nueces County into the new Anglo-majority-CVAP CD27. Thus, the Hispanics who had been in an opportunity district were no longer in such a district. 358. There is legally significant racially polarized voting in Nueces County. 359. Including all or most of the population of Nueces County in a southern district and in “the envelope” increases the Hispanic population available in the envelope to draw Latino opportunity districts, and makes it easier to draw HCVAP-majority districts CD20, CD23, and CD35. Including the population of Nueces County in the envelope makes it easier to draw seven Latino opportunity districts under § 2 without including Travis County. Congress - Travis County/CD25/CD35 360. In the benchmark Plan C100, Travis County was divided among three congressional districts, CD10, CD21, and CD25. TrA832 (Rodriguez); PL-1126 (map). Anglo growth in Travis County roughly kept pace with the minority growth in the area. TrA936 (Ansolabahere). CD25 had the smallest decline (-1.2%) of Anglo CVAP population of all the districts. TrA938 (Ansolabahere). 361. CD10 included northeastern Travis County/ Austin and extended all the way to western Harris County (Houston). The CD10 incumbent was Michael McCaul, an Anglo Republican. CD21 included western Travis County/Austin and extended to the southwest into Bexar County (San Antonio). The incumbent was Lamar Smith, an Anglo Republican. 301a 362. Benchmark CD25 was drawn by the court in 2006 after the Supreme Court’s ruling in LULAC v. Perry. LULAC v. Perry, 2:03cv354 (E.D. Tex. 2006); PL-230 (order adopting plan 1438C); PL-231 (opinion); D-648 (map). Benchmark CD25 included south and southeastern Travis County and a number of counties to the southeast. Anglo Democrat Lloyd Doggett won the special election for the new CD25 in December 2006, and won re-election in 2008 and 2010. Plaintiffs contend that CD25 was a functioning tri-ethnic coalition and/or crossover district. 363. More than half of the population of benchmark CD25 (486,125 persons) came from Travis County. The district was 49.8% Anglo in terms of total population and 54.9% Anglo in terms of VAP. Using 2005-2009 ACS data, CD25 was 63.1% Anglo CVAP, 25.3% HCVAP, and 9% BCVAP. Benchmark CD25 encompassed a vast majority of the African-American and Latino population and a majority of the progressive Anglo areas of the City of Austin. TrA832, TrA837 (Rodriguez); Tr1192 (Butts); Rod-922. 364. Sen. Gonzalo Barrientos testified that they established a progressive tri-ethnic coalition of Anglos, African Americans, and Latinos in Austin/Travis County in the early 1970s. TrA1165. Travis County was then majority-Anglo and remains so today. TrA819 (Rodriguez); TrA918 (Dukes). David Butts, a political consultant, testified that he helped create the Austin coalition and has been involved with it for 30 to 40 years. Tr1178. He testified that they recognized there could be a change in the Austin political climate and began to organize, sort of in tandem with the Hispanic and African-American communities. They would meet periodically to talk about the political situation in Travis County and the candidates. Tr1179 (Butts). 302a He testified that it has evolved because a number of political organizations have developed across the city, somewhat regionalized. They are not necessarily working in tandem, but will generally interview the same set of candidates, and endorse and promote the same slate of candidates. Tr1179-80. He testified that it is literally a slate, and includes a mixture of Anglo, Hispanic, and Black candidates being endorsed by the particular clubs. Tr1180. Although Anglos are about 50% of the population, these clubs are not all run or controlled by Anglos. Tr1181 (Butts). The clubs invite the candidates to come to a meeting of all the clubs and each club can vote to endorse candidates at that meeting. Tr1181-82 (Butts). Butts estimated that this slat- ing/endorsement process results in wins 80 or 90% of the time. TrA1182. He testified that candidates running countywide cannot win with just minority support; they need crossover support of Anglos to win. Tr1183. In some races the Anglo vote is split, and then the minority vote is decisive, such as in the race between Kathie Tovo and Randi Shade for city council. Id. Butts also testified that Hispanic and Black voters tend to vote together in the primary, unless they are running against each other. Tr1184. Butts cites the Gisela Triana victory over Anglo Jan Soifer in 2004 and the Sam Biscoe victory over Anglo Bob Honts in 2002 as examples of the coalition electing candidates. Tr1184-86. He also agreed that sometimes the coalition disagrees in the primary election; in the 2008 primary between Carlos Barrera and Anglo John Lipscombe the coalition split, and in 2010 African Americans supported Lipscombe and Hispanics supported Hispanic Olga Seelig. Tr1187-88. The evidence (including specifically exhibits Rod-901, Quesada-367, D-482, and D-486) shows that numerous minority candidates won partisan and nonpartisan elections in 303a Travis County between 2000 and 2010. Tr1188 (Butts). 365. Rep. Eddie Rodriguez lives in Austin, is Vice President of NIALC, and represents HD51, which is generally central east Austin, south central Austin, and southeast Travis County. TrA810, TrA846 (Rodriguez). Rodriguez described the tri-ethnic coalition as basically a coalition of African-Americans, Latinos, and progressive Anglos who tend to vote for the same candidate, the candidate that shares their values or agrees with them on certain issues. TrA817- 88. He opined that it happens in every general election and in the Democratic primary, as well as nonpartisan races such as city council and mayor. TrA818, TrA869. The coalition has worked historically to elect a Latino mayor (Gus Garcia) and African-American council members running city-wide. TrA869 (Rodriguez). Rodriguez stated that it played a role in his election in HD51, and that he gets the support of significant numbers of Anglos in his district, as well as African Americans and Latinos. TrA820-21. Rodriguez agreed that the tri-ethnic coalition is essentially made up of people who share similar political views. TrA861. He stated that to some degree the coalition includes clubs and organizations, including South Austin Democrats, Central Austin Democrats, Progressive Democrats, Capitol Area Democrats, Black Austin Democrats, Tejano Democrats, University Democrats, labor, and others, but mostly it is voters and the way they vote, and there is no formal decision making structure. TrA861-62, TrA868 (Rodriguez); TrA920-21 (Dukes). 366. Based on his familiarity with Travis County elections and politics, Rodriguez testified that African Americans and Hispanics generally support the same candidate in the Democratic primary. TrA821 304a (Rodriguez). In 2004, Hispanic Gisela Triana defeated Anglo Jan Soifer in the primary for district judge, and then won the general election. Similarly, African- American Greg Hamilton defeated Anglo opponents in the Democratic primary for Sheriff and won the general election. TrA823-24 (Rodriguez). In the 2008 race for County Tax Assessor-Collector, African- American Nelda Wells Spears defeated Anglo Glen Maxey by a large margin (74%). TrA826 (Rodriguez); D-412. Rodriguez testified that these were examples of the coalition working. TrA822-24, TrA827. Rodriguez agreed that there might sometimes be differences and supporting of different candidates, and that Anglo voters might sometimes support a different candidate than Hispanic and African-American voters. TrA862- 63. But they eventually come together and support the same candidate in the general election, usually the Democrat candidate. TrA863-64 (Rodriguez); TrA918 (Dukes). 367. Sam Biscoe (African American) testified that when he was on the Travis County Commissioners Court (east and northeast Travis County) from 1988 to 1997, he had to appeal to a coalition of voters because the precinct was only about 20% African American. Tr1202. He knew he could not get elected with just African-American votes, so he put together a platform to appeal to Anglos and Hispanics too. Id. He felt he could run for County Judge in an Anglo-dominated county (that was only 10% African American) because the coalition strategy had worked and he still had the support of the coalition. Tr1203. He defeated an Anglo in the 1998 Democratic Primary with support from the coalition. Tr1204-05 (Biscoe). He testified that a candidate cannot win the Democratic primary in Travis County without a significant percentage of the Anglo vote. Tr1205. The coalition of Anglo, Hispanic, 305a African-American, and now Asian voters still exists. Tr1206 (Biscoe). 368. Contested Democratic primaries are not common in Travis County. TrA864 (Rodriguez). Of 33 primary elections between 2002 and 2010 in which minority candidates won and then won the general election, only eight had contested primaries. TrA919 (Dukes); D-486 (supporting exhibits at D-406 to D- 415). It is also uncommon in Travis County Demo- cratic primaries to have two minorities run against each other. TrA825 (Rodriguez). In 2006, African- American Eric Shepperd defeated Hispanic Elena Diaz in the primary for County Court at Law Judge and won the general election. TrA824 (Rodriguez); Rod-901. It was a very close race for an open seat. Shepperd won with 50.41% of the vote. TrA825 (Rodriguez); TrA920 (Dukes); D-410. The coalition was evenly split between these two candidates. TrA825 (Rodriguez). Hispanic voters supported Diaz and African-American voters preferred Shepperd (close to 100%). TrA978-79 (Ansolabahere). 369. Rep. Dawnna Dukes is an African-American Democrat representing HD46 in northeast Travis County. TrA871-72 (Dukes). She testified that countywide, the tri-ethnic coalition is an organization of progressive Anglos, Hispanics, African Americans, and Democratic organizations and other organizations (such as labor), who endorse candidates to run for office and then work to assist them in becoming elected. TrA876-77, TrA918. Austin and Travis County readily support African-American and Hispanic candidates countywide; there is no label placed upon a person based on the color of their skin. TrA877 (Dukes). She testified that African Americans and Hispanics have 306a been elected countywide, and each group (Anglos, His- panics, and African Americans) wins about 1/3 of the time. Id. Dukes testified that the coalition supported Senator Barrientos against an Anglo opponent, that he won with 53% of the vote, and that he could not have won without Anglo support. TrA878-80. She testified that she won her primary in 2008 against an Anglo opponent with support from African Americans, Hispanics, and Anglos. TrA875. 370. Jeffrey Travillion of the NAACP and a resident of Pflugerville in Travis County testified that the different racial groups worked together to draw county commissioner precincts and that they work in cooperation to elect county commissioners. TrA1023-26. 371. Dr. Ansolabahere found that benchmark CD25 was a minority opportunity district (a functioning crossover district) due to the high rate of Anglo crossover voting and minority cohesion. TrA954 (reports at Rod-912 and Rod-914). He defined a crossover district as one that is majority white, but where there is sufficiently low white voting cohesion or sufficient crossover voting that minorities can elect their preferred candidates. TrA951. It is not just a Democrat or one- party district because a majority of whites are not voting with the minorities. TrA951-52 (Ansolabahere). He looked at endogenous general election results in CD25 to determine whether it was a crossover district. TrA953. He found that the minority-preferred candidate was Lloyd Doggett in general elections, and that he won with minority and Anglo crossover votes. TrA954, TrA969 (discussing Rod-912 at 30-31, Table 8 and Rod-914 at 20-23). 372. Dr. Ansolabahere also looked at primary elections and found that Doggett was the Hispanic- preferred candidate in the Travis County portions of 307a CD25 in the 2004 Democratic Primary when Doggett was challenged by Hispanic Leticia Hinojosa. TrA1012 (report at Rod-909).21 He further found that Doggett was the Hispanic-preferred candidate (i.e., the majority of Hispanics voted for Doggett) district-wide and in the Travis County portions of new CD35 in the 2012 primary. In that election, Doggett had two Hispanic opponents (Maria Luisa Alvarado and Sylvia Romo) and the district was a majority-Hispanic district constructed to be favorable to Hispanic voters. TrA983 (Ansolabahere). District-wide, Doggett received 53% of Hispanic votes, and was the Latino-preferred candidate. TrA983-84 (Ansolabahere). In addition, Hispanic (94%), African-American (92%), and Anglo (92%) voters in Travis County were highly cohesive in support of Doggett, while Romo and Alvarado received less than 5% of each of their votes. Rod-917. Dr. Ansolabahere notes that this is “indisputable evidence of a strong tri- ethnic coalition at work in Travis County.” Rod-917 at 5. He concluded that CD25 was a minority ability-to- elect district. Rod-914/PL-1142 at 16. 373. David Butts testified that Doggett has the support of the coalition, that he carried the minority vote by 90 or 95%, and that he would have been defeated in 2010 but for the minority vote. Tr1192-93.

21 In 2004, CD25 was a majority-Hispanic district drawn by the Texas Legislature; it was different from the benchmark CD25 because the court redrew CD25 in 2006 following LULAC v. Perry. Dr. Ansolabahere looked only at the Travis County portion of CD25, which overlaps substantially in the 2004 and 2006 versions. TrA981; Rod-909 at 2. He did not look at the 2006 special election primary because Doggett’s opponents were not Hispanic. Doggett was uncontested in the 2008 and 2010 primaries, so these do not provide cohesion information. TrA982, TrA993. Therefore, no analysis considers a primary election in the actual CD25 benchmark configuration. TrA993-95 (Ansolabahere). 308a Dukes also testified that Doggett was elected in benchmark CD25 with the support of Black and Hispanic voters, along with like-minded Anglo voters, and that he could not have been elected without African-American and Hispanic support. TrA893. She also testified that Doggett is the preferred minority candidate in the new CD35. TrA893. Rodriguez agreed that in benchmark CD25, Hispanics and African Americans were able to elect their candidate of choice for Congress, and that choice was Doggett. TrA843. 374. Dr. Ansolabahere testified that Travis County has the lowest level of Anglo voter cohesion in the State. TrA943. He found that typically 40 to 45% of Anglo voters will vote for the minority-preferred candidate. TrA943; Tr1120 (homogenous white precincts were voting 53% for Republicans, a noticeably lower level of cohesion than elsewhere in Texas); Joint Expert Ex. E-15 (Ansolabahere report) at 33. Anglo voters in Travis County do not vote as a bloc against minority-preferred candidates. Id. at 34; Tr1121 (Ansolabahere). 375. Dr. Ansolabahere testified that a crossover district was defined originally in terms of general elections, but a key question is also whether the primary elections serve as a filter to prevent the minority- preferred candidate from emerging (i.e., the minority- preferred candidate is losing to the Anglo-preferred candidate). TrA954-57. He criticized Dr. Alford’s and Dr. Engstrom’s analyses as just focusing on polarization and not looking at outcomes; he noted that their polarization analyses are helpful to understand part of the question (which candidates are preferred) but do not determine whether those candidates are winning. TrA957-58, TrA1010, TrA971 (he does not disagree with Dr. Engstrom’s cohesion statistics). He therefore 309a examined the success rate of minority-preferred candidates in elections in CD25 in Travis County and found that 73% of the time the candidate preferred by the racial minorities prevailed. TrA958; Rod-914 at 23. For each racial group (Anglo, Hispanic, and African American), their preferred candidate won about 1/3 of the time. TrA959. Therefore, he found that in Travis County, the primary does not serve as a barrier to the minority-preferred candidate. Id. He also found that to some degree, they all preferred the same candidate, and in all but six elections at least two of the groups preferred the same candidate. TrA960, TrA998; Rod-914 attachment 6. Specifically, of 43 primaries between 2006 and 2008,22 all three groups supported the same winning candidate in 14 elections; whites and blacks supported the same winning candidate in eight; white and Hispanics supported the same winning candidate in eight; and Blacks and Hispanics supported the same winning candidate in seven. TrA959-60. 376. Although African-American and Hispanic voters did not support the same candidate in the 2006 primary between Shepperd and Diaz (i.e., they did not vote cohesively), a minority-preferred candidate won, and thus the primary did not function as a screen.

22 The elections are listed at Rod-914 at 51-54. TrA974. They are not all county-wide primary elections. TrA974, TrA999. The number of precincts involved in each of the elections varied. TrA1002. Ansolabahere did not look at whether these elections involved a contest between a minority and non-minority candidate. TrA974, TrA1003. He did not view the race of the candidate as relevant to the determination of whether the candidate was preferred by minority voters. TrA977. Therefore, his analysis did not result in any opinion about how often Anglo or African- American voters support the Latino-preferred candidate in a racially contested Democratic primary. TrA1005. 310a TrA980, TrA1011 (Ansolabahere). Ansolabahere acknowledged that Hispanics, African Americans, and Anglos do not always vote the same way for every candidate in every race, and he included the Shepperd/ Diaz race in his 43- primary analysis. TrA980, TrA1011-12. He noted that the candidate preferred by Whites alone won only one of the primaries, whereas African Ameri- cans won two such races, and Hispanics won three. TrA960. Overall, Dr. Ansolabahere’s analysis of Dem- ocratic primaries showed that Hispanic-preferred and Black-preferred candidates win the majority of votes in the primary elections in CD25 and Travis County as often as White-preferred candidates, and there is some degree to which the three groups are actually coalescing inside the primaries. TrA960; Rod-914/PL- 1142 at 20. In 84% of the primaries studied, Whites preferred the same candidates as Hispanics, Blacks, or both. Rod-914/PL-1142 at 20. However, his report also shows that Blacks and Hispanics are not highly cohesive in the primaries, supporting the same candidate in 21/43 of the primaries. Rod-914, att. 6. 377. In his August 9, 2011 report on racially polarized voting, Dr. Engstrom noted that there is more support for Latino candidates by non-Latino voters in Travis County than other counties. In the bivariate analysis, while non-Latino voters did not provide majority support for the two Latino candidates with the Democratic Party nomination in the 2008 general election, they did so for the three in the 2010 general election. He noted that the multivariate analysis reveals this may be due to African-American voters, who support Democrat Latino candidates at levels above 90%, while the support among other voters (Anglos) for Latino Democrat candidates ranged from 41.4% to 48.1%. Engstrom 8-9-11 report, Joint Expert 311a Ex-7 at 16. Dr. Engstrom looked at the 2008 Demo- cratic Primaries for U.S. Senate and Supreme Court Places 7 and 8 and the 2010 Democratic primaries for Lt. Governor and Land Commissioner. Non-Latinos provided a majority of their votes to two of the five Latino candidates that were preferred by Latinos in the Democrat primaries studied. In the multivariate analysis, African Americans did not provide majority support for any of the Latino candidates in the Democrat primaries, while other voters (Anglos) did in two of them. Dr. Engstrom concluded, “Latino voters in Travis County have provided cohesive support for Latino candidates in both general elections and Democrat primaries. While African Americans have shared this preference in general elections, neither they nor other voters [Anglos] have been consistent supporters of those preferences in Democrat primaries.” Joint Expert Ex. E-7 at 17. 378. In his Corrected Rebuttal Report (docket no. 307-1), Dr. Engstrom analyzed 17 general and primary elections with Latino candidates within Travis County. He found that Latino voters were highly cohesive in support of Latino candidates with the Democrat Party nomination in the general election and in Democratic primaries. He looked at the 2006 Democratic Primary and runoff for Lt. Governor, the 2008 Democratic Primary for U.S. Senate, Supreme Court Place 7, and Supreme Court Place 8, the 2010 Democratic Primary for Lt. Governor and Land Commissioner, and the 2010 Republican Primary for Governor and Railroad Commissioner. Five of the six Latino candidates in the Democratic primaries received majority support from Latino voters in both the bivariate and multivariate analyses. In the bivariate analysis, non-Latino voters provided majority support for only two of these candidates. In the multivariate analysis, African-American 312a voters did not favor any of the Latino candidates, whereas other (Anglo) voters provided majority support for two of them. Engstrom Corr. Rebuttal Report (docket no. 307-1) at 21-24. He found that African Americans were strongly supportive of the Latino candidates with Democratic nominations in the general elections, but that support was not present for Latino candidates in the Democratic primaries, and that the “other voters” (Anglos) more often than not cast their ballots for the opponents of Latino-preferred candidates. Id. at 24. 379. Dr. Alford discussed Dr. Engstrom’s results, noting that in general elections, Hispanics, Blacks, and Anglos support the Hispanic candidates essentially the same as they do in other partisan elections elsewhere in the state. He states, The only thing that is distinctive is that in the general election, since there is a sufficiently large population of Anglos in Travis County that are Democrats, the Anglo vote for His- panic candidates that are Democrats is more evenly split than is the case elsewhere in the state. In other words, the behavior of voters in Travis County in general elections com- pares similarly [to] voters elsewhere in Texas - the key difference is the relatively large proportion of Democrats among Anglos in Travis County. If there is anything unique in the primary elections in Travis County, it is that Travis is the only county in the analysis where Hispanic cohesion falls below 70%. The average Anglo support for Hispanic candidates in Travis County, at 45%, is slightly above the average of 43%, but it’s below the 49% support in Bexar County. Black support 313a for Hispanic candidates, at 30%, is the lowest in any of the counties in the analysis and is well below the average of 40%. Alford 8-22-11 Report at 18-19. He also notes that “[t]he overall conclusion that Blacks and Hispanics are not cohesive in the primary elections is also the conclusion reached by Dr. Engstrom.” Id. at 19. Dr. Ansolabahere responds that Dr. Alford offers no evidence or analysis to support his assertion that “the distinctive feature of the Travis County electorate is that there is a high fraction of Democrats among whites.” Rod-913 at 4. Dr. Ansolabahere reiterates his prior assessment “that Blacks and Hispanics exhibit a high rate of cohesion because in most primary elections large majorities of these two groups vote for the same candidate, and the candidates preferred by minorities won 75 percent of the time in Travis County primaries.” Rod-913 at 4. 380. Dr. Ansolabahere opined that Dr. Engstrom’s results provide extensive evidence consistent with his results, showing high levels of cohesion among His- panic and Black voters and division among Whites, with typically 40 to 50% supporting minority-preferred candidates. Rod-914/PL-1142 at 21. Considering only Travis County, he noted that Blacks and Hispanics succeeded in electing their preferred candidates in eight of the nine elections examined, and that “Blacks and Hispanics routinely elect their preferred candidates in CD25 in Plan C100 and in the Travis County area generally.” Id. Looking at primaries, he found “a very high success rate of minority-preferred candidates inside the Democratic primary,” with minority- preferred candidates succeeding as often or more often than White-preferred candidates in the Democratic primaries in Travis County. Id. at 23. He also found 314a that no one group dominates the primary process; rather, power is shared very equally. Of the 43 minority primary winners, Whites backed 31; Hispanics backed 32; and Blacks backed 31. Id. Thus, he concluded, “it is clear that this is an area where minority- preferred candidates have the ability to win the Democratic primaries.” Id. at 23-24. 381. In his March 2014 Supplemental Report, Dr. Alford states, “In the primary contests, Dr. Engstrom’s results from his earlier report for Travis County establish that the tri-ethnic coalition is not a group of individuals who vote cohesively.” D-168 at 22. He essentially repeats his 8-22-11 Report conclusions, but adds “the voters in Travis County are more polarized than any other urban county in the State.” Id. at 23. Dr. Ansolabahere notes that Dr. Alford cites no facts to support this statement. He states that his reports show that Travis County has the lowest racial polarization of the most populous counties in general elections because of the high level of Anglo crossover voting. 382. At the field hearings, the Austin community desired representation in a single congressional district. US-340. At the February 16, 2011 hearing before the SRC, Sen. Kirk Watson noted that the 2003 congressional redistricting had divided some of the minority community in Travis County in a way that negatively impacted its ability to build coalitions, and he encouraged the committee to look carefully at assuring the coalitions could be put back intact and that nothing was done to further decrease the ability to have an effective vote. D-589 at 4. Rep. Dukes testified that there is a coalition between the Hispanic community and the African-American community and asked that they be kept together in a congressional 315a district. Id. at 5. Rep. Rodriguez, Mayor Pro Tem Mike Martinez, Travis County Commissioner Ron Davis, Travis County Sheriff Greg Hamilton, Travis County Attorney David Escamilla, and political consultant Alfred Stanley also testified about the coalition of voters in Travis County. Id. at 14-18. 383. At the April 7 HRC public hearing on the congressional map, MALDEF presented Plan C108, which included a new Hispanic opportunity district (CD28) that went from Bexar County to Travis County, stopping at approximately 7th Street in Austin. D-594 at 92-93. Later in the regular session, MALDEF submitted Plan C122 that also included CD28, a new Hispanic-majority-CVAP (52.3%) district that joined portions of Bexar County/San Antonio with southeast Travis County/Austin (stopping at approximately 12th Street in Austin). Joint Map Ex. J-3; D-643; Quesada- 382; TrA1759 (Downton). The district was based in Bexar County, including approximately 399,000 persons in Bexar County, and joined them with approximately 148,000 persons in Travis County. Joint Map Ex. J-3. It also included portions of Guadalupe (91,000 persons), Caldwell (25,000 persons), and Hays Counties (35,000 persons). The district had an area rubber band score of .546 and a perimeter-to-area score of .135. Joint Map Ex. J-3. The proposed district was 52.3% HCVAP using 2005-2009 ACS data and 53.3% HCVAP using 2008-2012 ACS data. D-546.6. Rep. Eddie Rodriguez (Hispanic, Democrat) from Austin was not in favor of MALDEF’s proposed district because he was concerned that it would affect the tri-ethnic coalition in Travis County. TrA850. 384. When Downton was drawing the map, he based the new majority-HCVAP district CD35 on MALDEF’s proposal. Tr916, TrA1673, TrA1758 (Downton). Downton 316a testified that he had been directed to create a new HCVAP-majority district, and when he realized they could draw the district like the one proposed by MALDEF, he focused on making that the new Hispanic- majority district. Tr916. Downton was not sure if CD35 was legally required; he said people disagreed. Tr987 (Downton); Downton 8-12-11 depo. at 114. He thought there was a potential risk if they did not create an additional Latino opportunity district. Tr951 (Downton). He did not think it necessary for § 5, but thought a court might find it was required by § 2. Tr988 (Downton). 385. On May 30, Lamar Smith emailed Opiela asking for Downton’s email and asking about CD23. D-613. He also said, “Also didn’t realize I had part of guadalupe.” Id. Opiela then responded and copied Interiano: “I didn’t think Lamar had Guadalupe . . . but it’s in the system printout is this a mistake, Gerardo? It’s only 13k people.” Id. Interiano wrote, “I don’t think that it was a mistake. I think this was done in order to make the VRA district work. But I can double check with Ryan tomorrow morning.” Id. 386. In the first public plan released on May 31, Plan C125, Travis County was divided among six congressional districts. Downton drew the Travis County configuration. Downton 8-12-11 depo. at 143. He testified that he started with the concept of the Bexar County to Travis County minority district proposed by MALDEF, with input from Bexar County Reps. Villarreal and Castro concerning the Bexar County portion of the district and input from Congressmen Smith and McCaul concerning the Travis County portion. Id. at 141, 144. This became the new HCVAP- majority district CD35. In Plan C125, the proposed CD35 had approximately 315,000 persons from Bexar 317a County and 202,000 from Travis County. D-548.2. It also had approximately 52,000 persons from Guadalupe County. D-548.2. It was 52.8% HCVAP using 2005- 2009 ACS data and 52.6% HCVAP using 2008-2012 ACS data. D-548.3; D-548.5 District election analysis showed President Obama receiving 43.6% of the vote, Noriega receiving 42.8%, Yanez receiving 44.6% of the vote for Supreme Court Place 8, and Molina 43.7% of the vote for Court of Criminal Appeals in the 2008 general election. D-687. Downton considered CD35 to be a Hispanic-majority opportunity district because it was over 50% HCVAP. Downton 8-12-11 depo. at 104. 387. Downton testified that after creating CD35, which extended into Travis County, he had to have a conduit from Congressman McCaul’s home in northern Travis County to the rest of CD10, which extends to Harris County. He also testified that Congressman Carter, whose district was based in Williamson County, refused to come into Travis County and effectively lobbied the Senate so that he would not, leaving Flores’s CD17 to come into “the place that’s left.” Downton 8-12-11 depo. at 141, 145. CD17 takes a sliver of northern Travis County and then includes a number of rural counties to the northeast. Downton testified that CD25, which had been a Democrat district represented by Lloyd Doggett, was intentionally drawn to be a Republican district, and includes the largest part of Travis County (including downtown, the Capitol, and UT). Id. at 142- 44; TrA1785. Doggett’s home, which is in a historically African-American neighborhood, was placed in the new CD25 in C185. TrA844 (Rodriguez). CD25 begins in Hays County south of Travis County/Austin, and then stretches up north to the Tarrant County line. 318a 388. Downton testified that he used partisan shading in Travis County to divide the Anglo Democrats among the Republican districts “for the partisan purpose of electing Republicans out of those districts.” TrA1674. He testified that he lives in Travis County and was aware that the southeast part of Travis County is generally Hispanic. That area became the Travis County portion of the new CD35. Downton testified that, at some point in trying to get CD35 over the 50% threshold, he turned on racial shading for Travis County to find the concentrated Hispanic populations to draw them in to get over the 50% benchmark. TrA1674-75 (Downton). He testified that he thought the use of racial shading was necessary to comply with § 2. TrA1675. When asked about inclusion of the “squiggle” in the north part of CD35 (in north central Austin, not in southeast Austin/Travis County), he stated that the area was included to pick up His- panic population. Tr989 (“If that is a Hispanic area, and I think it is likely that it is, then, yes, it would have been included in District 35 as we were trying to create a Hispanic majority district.”); Downton 8-12- 11 depo. at 121 (“There’s a Hispanic community there and we were trying to keep all of the Hispanic communities of Travis County together in one district to the extent possible.”). That is confirmed by racial shading; the “squiggle” picks up some 90-100% His- panic population areas. Downton testified that he tried to keep the “Hispanic community of interest in east Austin” together in CD35. TrA1776-77. Downton also testified that the district is narrow in the northeastern part of Bexar County because it “has a very low concentration of Hispanics, so if we had widened that out up there, it would cause a problem in keeping another majority Hispanic district.” Downton 8-12-11 depo. at 120. 319a 389. Butts testified that CD35 “basically takes in Hispanics and is very artfully drawn. It takes and cuts off a part of the black east Austin to basically connect it to this part, which is Hispanic and some black right there. So basically the African American community is split into four congressional districts. There’s a hundred thousand blacks in Travis County and Hispanics are basically split primarily. The heaviest concentration is [divided] into three congressional districts.” Tr1194-95. Precincts are split to place minority population into CD35. Tr1195-96 (Butts). St. Edward’s University, a Catholic school with a significant number of Hispanic students, is in a split precinct, with the administration building in CD21 and the dorms and student housing (more likely to be Hispanic) in CD35. Tr1198 (Butts). In the Precinct 4 split, Hispanics are concentrated in the point of the “arrowhead,” which is in CD35, and areas outside the arrowhead are in CD21. Tr1199 (Butts). 390. Downton agreed that he did not keep the African-American population together, with a portion of the concentrated African-American population in CD35 separated from the rest of the nearby African- American population in CD25. TrA1778-79. Downton testified that he had to put that African-American population in CD35 “to create a conduit to pick up the rest of the Hispanic population in the northwest part of 35.” TrA1779. As noted, Downton joined this population because it was Hispanic and he was trying to create a majority-HCVAP district. Id. 391. Downton testified that he applied certain traditional redistricting principles in Travis County, including redistricting for partisan purposes, protecting incumbents (Flores, McCaul, and Smith, but not Doggett), and compliance with the VRA with respect 320a to CD35.23 Downton admitted that he paid no attention to City of Austin boundaries. TrA1782. The district lines do not match up with any city boundaries, with House districts (TrA1780-81 (Downton)), or with any recognizable communities other than race. The Travis County district boundaries do not respect communities of interest. Partisanship and drawing the new CD35 as a majority-Hispanic district were the only redistricting principles being applied with regard to Travis County. Downton 8-12-11 depo. at 136. 392. When Rodriguez first saw Plan C125, he thought it was the worst-case scenario for Travis County. TrA827. He thought it was a direct assault on a successful coalition of minority voters and Anglo voters in Travis County. TrA827-28. He also did not like CD35, with Austin being paired with Bexar County/ San Antonio where San Antonio would anchor the district. TrA828.

23 Flores’s district CD17 had not been in Travis County before, and no district had ever joined east Austin to Forth Worth. TrA1781 (Downton). The new CD35 was not the first time a district had joined portions of Travis and Bexar Counties. CD21 includes portions of both San Antonio and Austin. Tr943 (Downton). PL-305 shows that for the 1996 special and general elections, and the 1998, 2000 elections, San Antonio and Williamson County are connected in CD21. Tr944 (Downton). And in the court-ordered map used for the 2002 elections, San Antonio and Austin are both in CD21. Tr944 (Downton); PL-306. Also in the legislatively drawn maps for the 2004 elections and 2006 primaries, they are united in CD21. Tr944 (Downton). However, CD21 was not a minority opportunity district and thus no court has considered whether it joins communities of interest in Austin and San Antonio. The same is true for CD10, which joined Austin and Houston in the benchmark. The fact that other districts may have joined distant cities does not support a finding that CD35 joins communities of interest in Austin and San Antonio. 321a 393. Sen. Gonzalo Barrientos, Sen. Joe Bernal, and Celeste Villarreal testified in favor a district joining Hispanic populations in San Antonio with Hispanics in southeast Austin. Barrientos testified that when he was a senator, he worked on issues that were relevant to Latino communities in both Travis County and Bexar County, including education. TrA1160. He testified that the San Antonio and Austin Hispanics in CD35 are only 60 miles apart and there is not much difference in terms of interests, music, culture. Id. He noted that some Mexican-Americans have familial connections between Bexar County and Travis County. TrA1160-61 (Barrientos). He generally supported the drawing of the new CD35 in 2011 and believes that CD35 connects communities of interest. TrA1168 (Barrientos). 394. Sen. Joe Bernal is from San Antonio, but he lived in Austin for eight years when serving in the Legislature. He testified that there are a lot of similarities between San Antonio’s south side/west side and southeast Austin, and they face a lot of the same issues with housing, income, and education. Tr557-58. Bernal felt that there was sufficient commonality of interests to justify combining the south/west side of San Antonio and southeast Austin in a proposed district, and that people in those parts of San Antonio have more in common with people in southeast Austin than with people in more affluent areas of San Antonio such as Alamo Heights (in CD21). Tr558-59 (Bernal). Celeste Villarreal saw more similarities than differences between San Antonio and Austin, and favored the district. TrA1134. 395. Rep. Dukes testified that she first saw the map the Friday before the Tuesday debate and was aghast with the configuration of Travis County, specifically 322a that it was further separated into five congressional districts. TrA880. She noted that, of the urban counties, Travis County is the only one that does not have a seat anchored within the county even though there is more than enough population to accomplish it. TrA881. She further testified that they separated the African American and Hispanic community but left the Republican Anglo community intact completely. Id. 396. Rep. Rodriguez, Rep. Dukes, and others worked together to see what they could do to keep Travis County as whole as possible and to maintain the coalition. TrA829, TrA858 (Rodriguez). They worked with the Travis County delegation and staff, Democratic party members in Travis County, and Joe Hamill from Lloyd Doggett’s office. TrA830 (Rodriguez); TrA888, TrA917 (Dukes). Dukes later sponsored Plan C166 in the House. TrA831. Plan C166 only created 7 majority- HCVAP districts, but CD25 was intended to and would function for minorities. TrA853, TrA870 (Dukes). 397. On June 1, Opiela forwarded Interiano an email with “summary stats” for Plan C125 from Mike Baselice. PL-311 Pt. 5; D-614; NAACP-618. Opiela also emailed Downton asking if he could swap certain precincts between CD10 and CD25 to “equalize GOP%” between CD10 and CD25. He wrote, “CD25 will still be over Statewide ORVS but McCaul won’t be floating below 55 McCain (I know RedAppl has him technically above, but that’s because RedAppl doesn’t take into account turnout when splitting VTDs)).” 398. At the June 2 HRC hearing, Solomons laid out Plan C125 and stated, “And I would regret to tell you that not all congressmen provided me with proposed maps for their districts, including the Honorable Lloyd Doggett.” D-601 at 8. Rep. Naishtat (Anglo, Democrat) of Austin spoke and asked that as much of Travis 323a County be placed in one district as possible. Id. at 39. He noted that it did not serve Austin or the rural populations that Austin was joined with to have them in districts together. Id. at 39-40. He further noted that Austin and Travis County regularly elect minorities and “[i]t does not serve our minority communities to be linked with representation outside of Austin; for example, with minority communities in San Antonio.” Id. at 40. He commented, “It’s outrageous that to this day the University of Texas, in the heart of my district; the state capitol, in the heart of my district . . . are represented by someone who does not live in Austin.” Id. at 40-41. Rep. Alvarado noted that the testimony from the field hearings in Austin about keeping as much of Austin and Travis County together as possible was not reflected in the map. Id. at 87. Deece Eckstein, the Intergovernmental Relations Officer for Travis County Texas, testified in opposition to the map and its treatment of Travis County. Id. at 134-36. At the time, Travis County was in six districts, with only four residents in CD31. Id. at 136. Solomons said they were trying to fix the four residents in CD31, but he knew what Eckstein meant with regard to the division. Id. at 138. Eckstein emphasized that Travis County voters did not come close to being 50% of any district and that Travis is the largest county in Texas and may well be the largest county in the country that does not have the opportunity to elect its own member of Congress. Id. at 141. Stewart Snider of the League of Women Voters of Austin testified against the map’s treatment of Travis County. Id. at 148-49. Gus Peña of Austin Concerned Hispanics also testified against the map’s treatment of Austin/Travis County. Id. at 150-54. Anne McAfee, Bill McClellan, David Thomas also testified against the map’s treatment of Austin. Id. at 157-58, 159-60, 160-62. In response to Mr. Thomas, 324a Solomons stated that Doggett had never contacted him. Thomas replied that he was not there on behalf of Doggett but on behalf of Austin voters. Id. at 162. Carl Lindemann, Bill Reid, Erin Foster, Stacy Suits, Ted Malina Raab, Bertha Means, and Mary Copenger testified against the plan based on its treatment of Austin/Travis County. Id. at 165-68, 189-91, 198-200, 217-19, 219-21, 221-23. 399. Between Plan C125 and Plan C130, a few changes were made between the border of CD20 and CD35 in San Antonio, and changes were made to CD17, CD10, and CD25 within Travis County. The revised map split Travis County among only five districts instead of six (by fixing the small encroachment of CD31 into Travis County). The Senate Committee Plan C136 and the Senate engrossment Plan C141 incorporated those changes 400. At the June 3 hearing before the SRC, when Seliger was asked which Congresspersons had input in the map, he stated that Flores, Doggett, McCaul, and Smith all had spoken with him or come to the Capitol. D-602 at 15. Sen. Watson (Anglo, Democrat) testified in opposition to C125 because of its treatment of Travis County, which was “sliced into five different congressional districts” and the City of Austin, which was divided into six different districts, despite its population being sufficient to have almost two whole districts of its own. Id. at 32. He noted that Travis County would not make up more than 35% of any district and that the treatment of Travis County was inconsistent with that of other large counties. Id. at 33. He also complained that the minority population of Travis County had been divided and that the map ignored their historically effective coalition. Id. at 33. He further noted that the only district in which Travis 325a County voters were a plurality of the district was CD25, which was 67.3% Anglo, and thus Anglo voters were being treated better. Id. at 35. Watson further explained that the map did not respect communities of interest, especially counties. Id. at 37. Watson noted that CD25 comes into east Austin and picks up Doggett’s home and a lot of the minority community of Eastern Travis County and puts them in a district that runs to the suburbs of Fort Worth. Id. at 39. Watson testified that C125 was retrogressive because it decreased the number of districts in which minorities could elect candidates of choice. Id. at 40. Stacy Suits opposed the treatment of Travis County and asked that CD21 and CD35 not come into Travis County and that CD25 be primarily drawn in Travis County. Id. at 53. Sandra Seekamp testified against the map’s treatment of Austin/Travis County. Id. at 74. Nina Perales stated, “I can say that we had a full and fair opportunity to communicate our views to the Chair- man of both the Senate and the House Redistricting Committees” and noted that although the concept of CD35 was taken from the MALDEF plans (C122), C130 had a different configuration than what MALDEF proposed. Id. at 80. Kunda Wicce and Maria Jimenez testified against the map’s treatment of Travis County/ Austin. Id. at 89-90. Stewart Snider testified against C125 and C130 for denying Travis County a congressional voice. Id. at 100. Eckstein again testified in opposition to Plans C125 and C130. Id. at 116. He read a resolution into the record, which asserted that the district lines destroyed a functioning crossover district. Id. He also complained about the way Travis County was split so that it was only a small part of five districts, and that this was inconsistent with the way other counties were treated. Id. at 117. 326a 401. When Sen. West asked Seliger why Travis County was divided that way, Seliger replied, “It was just, simply, as we went through the map, and, and put together the right size of districts, just did it that way.” D-602 at 118. Sen. West asked, “It just happened that way?” and Seliger answered, “Yes, as we went in from the corners of the edges of the map and we came in to get the number of people to get 698,488.” Id. Sen. West then asked if Committee counsel could be consulted about Travis County, and Morrison, Guinn, and Heath stated they thought it was constitutional. They also stated that they had just seen the plan and that they had not been involved in drawing the plan. Id. at 126. West asked if they believed that the map would pass muster under § 5. Guinn stated that he would want to study past election results before saying whether the plan complied with § 5. Numerous witnesses testified against the plan’s treatment of Austin and Travis County and combining Austin and San Antonio in the new district. Id. at 142-43 (Denardis); Id. at 146-47 (Guthrie); Id. at 151 (Rumancik); Id. at 152 (Damude); Id. at 156 (Sieverman); Id. at 159 (McAfee); Id. at 159-60 (Sosa); Id. at 160 (Yanez); Id. at 166-68 (Cortez); Id. at 169-70 (Tabor); Id. at 172 (Sparks). 402. At the June 6 floor debate in the Senate, Sen. Watson again complained about the map’s treatment of Travis County, splitting it into five districts and “ignoring completely and totally the concept of compactness and communities of interest.” D-22 at A-28. He also stated that it discriminated against the minority community in Travis County that has worked to develop a coalition that elects candidates of choice. Id. at A-29. 327a 403. On June 8, Solomons released statewide substitute Plan C144, which made changes to the border between CD21 and CD35 and between CD10 and CD25 in Travis County. The plan maintained the configuration splitting Travis County into five congressional districts. Significant changes were made to CD35 in Plan C144. Some population (approximately 14,000 persons) was moved from CD21 to CD35 in Travis County. The corridor connecting the district in southern Guadalupe County was shifted from Guadalupe County to Comal County and thus the Guadalupe County population in CD35 was reduced. And numerous changes were made among the districts within Bexar County. 404. In the 2011 trial, Downton testified that the changes in the southern region of CD35 between Plan C125 and Plan C144 were made at the request of Rep. Villarreal and Rep. Castro, both Hispanic Democrats, who wanted it to be a Bexar County-based district.24 Tr916-17, Tr985. He testified that they asked him to move more of CD35 into Bexar County and into specific areas of central San Antonio/downtown. Tr985- 86. He also testified that they “wanted to see the His- panic percentages raised with Hispanics from Bexar County, so we tried to accommodate them.” Downton 8-12-11 depo. at 116. Between Plan C125 and C130, CD35 takes more of downtown San Antonio from CD20, and Plan C144 takes even more of downtown

24 In the 2014 trial, Downton’s testimony was that he met with Rep. Villarreal and Rep. Castro at the beginning of the process and they told him to draw the new district with two thirds of it in San Antonio, to allow it to elect a San Antonio representative and they told “us” certain parts of San Antonio that they wanted in it. TrA1673. He testified that when he first drafted the map, he put those parts in, along with the Hispanic areas of Travis County and a line connecting them. TrA1673-74. 328a San Antonio from CD20. However, the district was not weighted more heavily toward Bexar County in these maps. In Plan C144, 314,178 persons are in Bexar County (down from 314,994 in Plan C125) and 215,626 are in Travis County (up from 201,700). D-555.2. Downton likely meant the changes made in Plan C170, in which the Bexar County population of CD35 was increased from 314,178 to 329,237. D-567.2 The changes made in Plan C144 were incorporated into the House Committee report Plan C149 on June 9. No further changes were made to Travis County after June 9. 405. Solomons made further changes to CD35 with his West Texas Amendment Plan C170, which he introduced on the House floor on June 14. The northern part of the Guadalupe County portion of CD35 (about 15,000 persons) was moved from CD35 to CD15, further narrowing the corridor joining the northern and southern portions of CD35. Approximately 23,000 Guadalupe County residents remained in CD35 along I-35. And further changes were made to the districts in Bexar County. In Plan C170, the Travis County portions of CD35 are unchanged. The Bexar County population of CD35 was increased by about 15,000. Downton testified that the changes to the corridor in Guadalupe County (the narrowing) were made due to the combined request of Rep. Kuempel to keep Guadalupe County whole and the request of the San Antonio representatives to have more of the district in San Antonio. Tr991. However, Guadalupe County was not made completely whole, and the record indicates that the configuration of CD35 through Guadalupe County was likely “to make the VRA district work” since they had previously included portions of Guadalupe County in CD21 to “make the district work” and no other reason was given for overriding Kuempel’s wishes. D-613. Downton also testified that 329a he tried to keep as much of Guadalupe County whole (in CD15) as possible, except for a small city (Schertz) that they also tried to keep close to whole (in CD35). TrA1655, TrA1663. The fact that they did not keep Guadalupe County whole in CD15 or the City of Schertz whole (part is in CD15) is evidence that other redistricting principles were subordinated to race in the drawing of CD35. 406. On June 14, during debate in the House, Reps. Turner (African American, Democrat) and Y. Davis (African American, Democrat) offered Plan C155, the Texas Legislative Black Caucus plan, which included a new Hispanic district CD10 that joined minority populations in Travis County with minority populations in Bexar County. Travis County was in only two districts—CD25 and CD10. Rep. Naishtat (Anglo, Democrat) and Turner discussed the fact that minority communities in Travis County worked as a coalition, and that they did not want Travis County Hispanics carved out and placed in a district with San Antonio. D-23 at S18. Solomons opposed the map, citing “legal concerns,” including that its new Hispanic districts were below 50% SSVR and HCVAP. Id. at S19. Solomons moved to table, which passed 93 to 49. Id. at S19-S20; PL-218 at 6. 407. Rep. Martinez Fischer (Hispanic, Democrat) offered Plan C163, in which Travis County is divided between only two districts. Joint Map Ex. J-5. Solomons opposed, stating that it created the same number of Hispanic majority districts as his plan, did not reflect the input received “from a number of folks,” and “splits a number of cities.” D-23 at S23. Solomons moved to table. Martinez Fischer said Plan C163 created an additional SSVR 50% district, and pointed out that the committee map cuts cities like Austin and San 330a Antonio, so they should be consistent if that is the standard. Id. at S24. The amendment was tabled 92 to 48. Id. 408. Martinez Fischer offered Plan C164, which he described as another statewide map that aimed to draw a new Hispanic district without affecting anybody else’s political liability. D-23 at S26. This map drew a district from Bexar County to Bastrop County that did not cut into Travis County. Joint Map Ex. J- 6. Travis County is divided between only two districts. Solomons opposed and moved to table. D-23 at S26. The amendment was tabled. Id. at S288. Downton testified that the Plan C164 configuration would have necessitated significant changes elsewhere in the map and in his opinion was not any more legal, so it was not included. Downton 8-12-11 depo. at 137. Downton testified that he did not get demonstration maps that had more majority-HCVAP districts than the committee map and in order to pass a map through the House, they needed to increase the net number of Republican districts by three. Id. at 138-39. 409. Martinez Fischer then proposed his third plan, Plan C165, which “takes a similar approach by connecting Travis County and Bexar County to create that additional district. It also maintains the two districts for the Dallas metroplex area, the one minority opportunity district in Harris County, for a total of four. Even still, on the third rendition, it performs better statistically than the Seliger/Solomons map by taking one additional SSVR district, and it also performs better by taking the BHVAP seats and growing them from 13 in the Solomons/Seliger map and growing them to 15.” D-23 and S28. This map splits Travis County among three districts, including a district that joined southeast Travis County with portions of Bexar 331a County. Solomons stated, “Not to belabor the point, but it’s similar arguments as the last two maps, members, and I hope you’ll stick with me and move to table.” D-23 at S28. The amendment was tabled. D-23 at S30. 410. Dukes offered statewide substitute Plan C166. Although Dukes was concerned primarily with Travis County, to redraw that area required drawing a new statewide map. TrA898 (Dukes). Travis County has CD25 wholly within it (including the minority areas) and the rest is in CD10 joined with Harris County (similar to the benchmark). Dukes stated, “The Austin- to-San Antonio district, CD35 in the republican plan in SB4, was moved to the Valley to give the Valley the new seat that its growth called for, but as well to keep Austin’s community of interest intact. It is labeled [CD33].” D-23 at S30. She continued, “This map keeps the minority coalition of Hispanic and African Ameri- can voters in east Travis County together so their voice and vote can be effective. This is in stark contrast to SB 4’s purposeful discrimination of these minority communities. In SB4 Austin voters make up a unique community of interest with strong diversity and respect for different points of view – that’s why we’re considered the heart of Texas, the oasis of Texas. In Travis County, Hispanics and African Americans, and Anglos act as a coalition, are able to elect the candidate of their choice from all races. Nearly every Hispanic elected official in Travis County has signed a letter stating they do not want Hispanic families to be carved out of Travis County and connected with a distant population in San Antonio. . . . This plan works to ensure that African Americans, and Hispanics, and like-minded Anglo voters in eastern Travis County, eastern Austin, are able to continue a coalition and elect an individual who will properly represent their 332a voice without the creation of purposeful discrimination as that which is done in SB 4.” Id. at S30. Dukes stated that “they have purposefully gone so far in their attempts to eliminate one person [Lloyd Doggett] that they have eliminated the voice of 688,000 people in Travis County.” Id. at S32. Rep. Howard (Anglo, Democrat) pointed out that Travis County is the only county of the 12 largest that does not have even one congressional district in which Travis County would be more than 50% of the district population, depriving Travis County of substantive representation. Id. at S34. Rep. Naishtat (Anglo, Democrat) also spoke in favor of the Dukes plan, arguing that Austin and Travis County have elected minorities and deserve a united voice in Washington. Id. at S35. Solomons opposed, stating, “I have some legal concerns about the map. It creates one less Hispanic district – Hispanic majority district – than the committee map, and neither the new District 35 in North Texas nor District 36 in Harris County are Hispanic minority districts. In fact, this map also retrogresses District 29, and quite frankly Ms. Dukes[‘] map, as I say, creates one less majority district than the committee map.” Id. at S36. Dukes responded, “I’m not quite certain where Representative Solomons pulls his numbers and sometimes I wonder about the concern put forth on minority communities, when Congressional District 29 is used as the litmus test for what is considered retrogression when one arguably can state something different.” Id. at S37. She also stated that Travis County has enough voters for a congressional district entirely within Travis County lines, and that it did not seem like the drawer of C149 “was too concerned about minority community and retrogression” because western Travis County, where the majority population is Anglo, is “entirely in one district, but eastern Travis County, 333a where African Americans and Hispanics live, is drawn into [four] separate in-and-out congressional districts.” Id. at S37. She said, “That is purposefully drawing African Americans and Hispanics into smaller portions and separation to create a GOP plan of purposeful discrimination. This map is not about ensuring one member of congress does not have a voice in Washing- ton. It is about ensuring that African Americans and Hispanics from eastern Travis County, from east Austin, will be unable to have a candidate of their choice to represent them in Washington.” Id. at S-37- S38. The amendment was tabled. Id. at S39. 411. Rep. Dukes then argued again against Solomons’ plan, arguing that it was motivated by discrimination in Austin and Travis County and that the process did not allow for meaningful input. D-23 at S45-S46. Rep. Alonzo also argued that there were § 2 and § 5 violations. He asserted that Doggett was the candidate of choice in CD25 and that the mapdrawers willfully chose to reduce minority opportunity. Id. at S50-51. He entered an exhibit asserting that the plan unnecessarily makes dramatic changes to the existing opportunity districts, packs and cracks the minority population, and splits up the Travis County minority coalition. Id. at S54-S57. No changes were made to Travis County in response to any of the testimony. 412. Plan C185 divides Travis County among five congressional districts. Dr. Ansolabahere testified that Plan C185 does not follow the traditional redistricting principle of keeping counties whole with respect to Travis County, and that Travis County is the only large county that does not have a district anchored in it (meaning that 50% of a district’s population comes from the county), even though it has over 1 million people. TrA944. County Judge Sam Biscoe also noted 334a that in Plan C185, Travis County voters do not have an anchor congressman and would be competing against other counties for federal dollars. Tr1207. He noted that the minority population is divided, resulting in diminished strength of minority voters who now make up less of the district population. Tr1208. 413. HD51 in southeast Travis County, which had been wholly in benchmark CD25, is mostly in the new CD35 (including the bulk of the Latino population), with parts also in CD10 and CD21. TrA810, TrA834 (Rodriguez); Rod-920; Rod-921. The Hispanic population of Travis County is primarily in CD35, but some are also in CD10 and 17, and smaller amounts in CD21 and CD25. TrA839-40 (Rodriguez). The new CD35 takes in an area of Travis County that is 76% Black or Hispanic (and 65% Hispanic). Rod-914/PL-1142 at 28. In Plan C185, the dense African-American population of east Austin is divided between CD25 and CD35. TrA838-39 (Rodriguez); Rod-923. Much of the African- American population in the center of Austin is placed into a district (CD25) that extends northward all the way to the Tarrant County border and includes nine predominantly Anglo rural counties. Tr1194 (Butts); Rod-914/PL-1142 (Ansolabahere report) at 26. The African-American population of eastern Travis County is divided among four districts (CD10, CD17, CD25, and CD35) and is “fractured.” TrA1405 (Murray); TrA838-39 (Rodriguez); Tr1193-94 (Butts); Rod-923. 414. Rep. Dukes testified that African Americans in Austin were historically forced to live in the “Negro district,” east of I-35, which is now a compact area of historical black neighborhoods in east Austin. TrA883- 84. They were in benchmark CD25, and they supported Doggett. TrA884 (Dukes). She noted that in C185, some of the core black community was left in 335a CD25, but the new CD25 goes up to and is anchored near Fort Worth, no longer in association with a community of interest in Austin. TrA885. What would have been more of the old CD25 is now in the new CD35, and very little of that African-American community was placed into the new CD35 that goes to San Antonio. TrA885 (Dukes). In C185, Dukes’s house district is in northeast Travis County (which has tracked the African-American population) and is in five separate congressional districts. TrA887-88 (Dukes); Rod. Ex. 921. Murray testified that the African-American community in Travis County is pretty effectively neu- tered by C185 by connecting them up with folks they share little with in another metropolitan area. Tr1045. Dr. Ansolabahere agreed that the C185 configuration “will have serious adverse consequences for Black voters in Travis County.” Rod-914/PL-1142 at 28. In the benchmark, most Travis County African Americans live in a district in which they can elect their preferred candidate, but those placed into CD10, CD17, CD21, and CD25 (which district includes over half the Black population of Travis County) will not have that ability. TrA1406 (Murray); Rod-914/PL-1142 at 28-29. 415. Jeffrey Travillion testified that in Plan C185 the minority community is split into five districts, which diminishes their influence. TrA1027. He testified that the congressional district lines in Travis County do not respect communities of interest such as school attendance zones, and put African-American neighborhoods near each other into different districts. TrA1028-30 (Travillion). He testified that Colony Park, an enclave of large African-American population, is connected in CD25 to West Austin/West Travis County, which is affluent, and those communities do not interact or work together. TrA1031. 336a 416. Dr. Ansolabahere also opined that Plan C185 will decrease the ability of Hispanics in Travis County to elect their preferred candidates. Rod-914/PL-1142 at 29. Although Hispanics placed into CD35 will have the opportunity to elect, roughly 201,881 Travis County Hispanics are placed into districts lacking such opportunity. The net effect of Plan C185 compared to Plan C100 is to reduce the number of Travis County His- panics residing in districts where they can elect their preferred candidates by 63,000. Rod-914/PL-1142 at 29. 417. Dr. Ansolabahere testified that visual observation of district lines with racial shading suggested that the boundaries of the districts that divide Travis County tend to be dividing up the minority population. TrA945; Rod-923. He noted that the thirteen VTDs with the highest percentage BVAP were all in CD25 in the benchmark, except VTD105, which was in CD10, and that the enacted plan splits these VTDs across three districts (CD10 (3 VTDs), CD25 (6 VTDs), and CD35 (4 VTDs)). He also noted that the concentrated Hispanic population VTDs from CD25 and CD10 are put into the new CD35. Rod-912 at 35. Dr. Ansolabahere confirmed his visual observation by looking at whether the racial composition of a district (using VAP25) would predict a likelihood of which district it would be in, and he found a high correlation, indicating that the district lines were based on race. TrA945-47. He also looked at whether party vote (based on percent of the two-party vote won by Democrats in six elections) correlated with which district a VTD would be found in, and found that it was evident that Plan C185 divided

25 Dr. Ansolabahere testified that he used VAP because he used data from RedAppl, which does not have CVAP data, and did not think using CVAP instead of VAP would change the correlation coefficient in a meaningful way. TrA948-49, TrA1010. 337a benchmark CD25 more along racial lines than party lines. Rod-912 at 41. With the exception of CD10, racial factors are significant predictors of which VTDs ended up in which districts in Travis County. TrA947- 48, TrA988 (Ansolabahere); Rod-912 at 39. He also found a party vote correlation related to inclusion in CD25 and CD35, but not CD10 or CD21 (and only weakly for the 2010 Governor’s race in CD17). Rod-912 at 39. Race is a stronger predictor than party vote of which VTDs are put into which districts. TrA947 (Ansolabahere); Rod-912 at 39-40.26 418. With regard to CD35, race was a strong predictor of whether a VTD in Travis County ended up in the district. TrA988 (Ansolabahere). The higher percent Hispanic a VTD, the more likely it was included in CD35, and the higher percent Anglo a VTD, the less likely it was included in CD35.TrA988 (Ansolabahere). Race was a stronger predictor (more than twice) than political votes. TrA989 (Ansolabahere). This is consistent with Downton’s testimony that he used racial

26 This data is in Table 12 of his report. For CD35, racial correlations are quite strong (.73 and -.60 for % HVAP and % Anglo VAP respectively) and party correlations are present as well, but are only about half as strong. TrA948; Rod-912 at 39. For CD25, party correlation is present (.20 to .27) but not as strong as racial correlation (.34 and -.38 for %HVAP and %Anglo VAP respectively). TrA948; Rod-912 at 40. For CD21 the race correlations are .21 and -.24 for %Anglo VAP and %BVAP respectively, and party vote correlations were not significant. TrA947; Rod-912 at 40. For CD17, there was a significant correlation for %BVAP (.19) and for party vote in the 2010 Governor’s election (-.15). TrA947; Rod-912 at 40. There were no significant correlations for CD10. TrA947; Rod-912 at 40. Dr. Ansolabahere also looked at the correlation of race and politics for Travis County VTDs from former CD25 only (Table 13) and found an even stronger correlation for race and a weaker correlation for party vote. Rod-912 at 40. 338a shading at the block level when drawing CD35 in Travis County to find Hispanic population to include in CD35. TrA1674-75 (Downton). 419. Dr. Ansolabahere also looked for correlations associated with all VTDs (not just Travis County VTDs) from former CD25. TrA949; Rod-912 at 41-42, Table 14. He found that Plan C185 divided CD25 more along racial lines than party lines, noting that race is highly predictive of which VTDs end up in CD25 and CD35, and has significant correlations for CD10 and CD21. TrA949-50; Rod-912 at 41. For CD25, there was a strong racial correlation and no party correlation. TrA949; Rod-912 at 56. For CD35, there was a strong racial correlation (.73 for HVAP, -.65 for Anglo VAP)), and the party correlation was about half of the racial correlation (.31). TrA949 ; Rod-912 at 56. For CD21, there was a strong racial correlation, and some party correlation (about half of the racial correlation). TrA949; Rod-912 at 56. For CD27 and CD34 (which were all VTDs outside Travis County) there is a party correlation but little to no racial correlation. TrA949. On the whole, looking at all the districts that divided Travis County or all the districts that divided old CD25, Dr. Ansolabahere found that race is a stronger predictor of where the lines fell than party. TrA950. 420. In the new CD35, Anglo CVAP is 34.5%, Black alone CVAP is 11.1% (11.3% combined BCVAP), and HCVAP is 51.9% using 2005-2009 ACS data. CD35 in Plan C185 “cuts more than 100 existing voting precincts.” Joint Expert Ex. E-11 (Korbel report) at 8, (Korbel supp. report) at 3; Red-381 Report (CD35 splits 106 VTDs). These splits were not all necessary to achieve population equality. Joint Expert Ex. E-11 (Korbel supp. report) at 3. CD35 splits 26 precincts in Travis County, 41 precincts in Bexar County, 7 in 339a Caldwell County, 6 in Comal County, 6 in Guadalupe County, and 14 in Hays County. Red-381 Report. 421. CD35 has a perimeter to area score of .054 and an area rubber band score of .364. It is the least compact district in Plan C185 under both measures. Giberson report (E-18) at 6-7. Downton agreed that CD35 is borderline in terms of compactness, but felt it was compact enough for legal compliance. Tr988; Downton 8-12-11 depo. at 135. Downton stated that there were reasons for every part of the way CD35 was drawn—either political reasons or legal compliance reasons. Id. at 136. The evidence indicates that significant portions of the district were drawn on the basis of race, and that race subordinated other redistricting principles. All portions of Travis County were included solely on the basis of race without regard to communities of interest or other redistricting principles. Most or all of the parts of Bexar County that were included were chosen because of their high percentage of His- panic residents and voters (although it may have been a political aim to weight the district to Bexar County, those Bexar County voters that were included were included primarily because they were Hispanic). 422. Dr. Ansolabahere opined that the dismantling of CD25 is evidence of intentional discrimination because it reveals how there could be more than 3 million more Hispanics and Blacks in Texas but only one new district created. Rod-914/PL-1142 at 24. He felt that Plan C185 did not respect the political gains that minorities in Travis County had already realized. Id. at 25. 423. The legislatively drawn CD35 remains in place in the interim plan. In the 2012 primary in the new CD35, Travis County and Bexar County Hispanics were not strongly cohesive in support of Lloyd Doggett. 340a 94% of Travis County Hispanics supported Doggett, but only at 42% of Bexar County Hispanics did. Rod- 917. Ansolabahere opined that this is evidence that the two Hispanic communities do not have common interests inside the Democratic Primary. Travillion also felt that it would be hard for communities in Austin and San Antonio to work together given the distance and lack of common institutions and working together. TrA1041-42 (Travillion). 424. The Gingles § 2 preconditions are not met in Travis County. Ansolabahere 8-22-11 depo. at 33. Anglos do not vote as a bloc to defeat minority candidates, and there are many minority-supported minority candidates who win office. Anglo voters do not show high levels of cohesion and do not vote as a bloc against the minority-preferred candidates; Blacks and Hispanics have the ability to elect their preferred candidates precisely because Anglo voters vote at sufficiently high rates for minority-preferred candidates. Rod- 914 (Ansolabahere report) at 18. The State of Texas asserted in the D.C. Court litigation that minorities in benchmark CD25 do not vote cohesively and that Anglos do not vote as a bloc to defeat the minority- preferred candidate. Docket no. 218 at 2 (NAACP-4). Texas had no reasonable basis for believing that a § 2 minority opportunity district should be drawn in Travis County. 425. An additional Hispanic opportunity district can be drawn in the envelope without disrupting the CD25 crossover district (as was done in Plan C220). TrA961 (Ansolabahere). CD25 or CD35 is a false choice; the map can include a new CD35 and also leave Travis County as whole as possible. TrA842-43 (Rodriguez); TrA892 (Dukes). 341a 426. Travillion testified that NAACP Plan C193 (which creates a new Central Texas district that does not come into Travis County and keeps the Travis County minority population whole in its own district) and NAACP Plan C260 (a proposed Travis County district) keep communities of interest together, giving them an opportunity to work together, and giving them a voice with their congressional representative. TrA1032-34. Murray agreed that a coalition district largely made up of the areas east of I-35 was preferable to Plan C185 and was justified. TrA1406. 427. Travillion testified that the African-American community works very well with lots of the Latino community and parts of the progressive Anglo community, as well as the Asian community because the interests of their children are the same. TrA1021. He testified that in recent years the interaction between the African-American community and the Republican Party has not been good. He pointed to the Republican Party’s adoption of a platform position to eliminate the VRA, the support of voter ID, and the tenor of discussions, including those about President Obama. TrA1038. On the NAACP Federal Legislative Civil Rights Report Card 2009-2010, McCaul and Smith got an “F” while Doggett had 90%. TrA1036; Quesada-8. The 2013 report card shows McCaul, Flores, Smith, and Williams all with a grade of “F” and Doggett with an “A.” NAACP-58. The African-American community pays attention to the NAACP report card, which suggests that their interests are not valued at all by those getting an “F.” TrA1040 (Travillion). Travillion testified that Republicans are not responsive to the African-American community. TrA1036. The African- American community was offended by Texas Attorney General Abbott continuing to campaign with Ted Nugent after he called Obama a mongrel. TrA1040. 342a 428. In Travis County, the district lines of CD35 are drawn on the basis of race to include Hispanic population in the district, and the use of race subordi- nates other traditional redistricting principles. Downton did not include all Hispanics from Travis County in CD35 because he wanted it to be a Bexar County weighted district. However, those Hispanics that were included in CD35 were included on the basis of their race and no other basis. Most or all of the parts of Bexar County that were included were chosen because of their high percentage of Hispanic residents and voters. Downton intended to draw an HCVAP- majority district and this criteria could not be compro- mised for any other redistricting criteria. 429. The Gingles preconditions are not satisfied in Travis County, and the Legislature had no reasonable basis for believing that a § 2 Latino opportunity district should be drawn in Travis County. 430. It was not necessary to dismantle CD25 to draw a new Latino opportunity district. Mapdrawers used the creation of CD35 as an excuse for the destruc- tion of CD25. Because mapdrawers would only consider a map that resulted in a net gain of three Republican seats, to draw a new Latino opportunity district to replace CD27 (switched to an Anglo Repub- lican district to protect Farenthold) and draw a new Latino-majority district, they had to get rid of Doggett’s Democrat district. Congress - CD23 431. CD23 was the subject of litigation after the 2003 redistricting. On appeal, the Supreme Court found that the State violated § 2 of the VRA in its configuration of CD23 in West Texas. LULAC v. Perry, 548 U.S. 399, 427 (2006). The Court noted, “After the 2002 343a election, it became apparent that District 23 as then drawn had an increasingly powerful Latino population that threatened to oust the incumbent Republican, Henry Bonilla.” Id. at 423. The Supreme Court stated, “[I]t is evident that the second and third Gingles preconditions—cohesion among the minority group and bloc voting among the majority population—are present in District 23.” Id. at 427. It further found that “the Latino majority in old District 23 did possess electoral opportunity protected by § 2.” Id. Although the old CD23 was a Latino opportunity district, the new CD23 unquestionably was not. Id. at 428. The Court noted, “[T]he new plan divided Webb County and the city of Laredo, on the Mexican border, that formed the county’s population base. Webb County, which is 94% Latino, had previously rested entirely within District 23; under the new plan, nearly 100,000 people were shifted into neighboring District 28. The rest of the county, approximately 93,000 people, remained in District 23. To replace the numbers District 23 lost, the State added voters in counties comprising a largely Anglo, Republican area in central Texas. In the newly drawn district, the Latino share of the citizen voting-age population dropped to 46% [from above 50%].” Id. at 424. Texas argued that it offset the loss of CD23 by creating the new CD25, but the Supreme Court concluded that the new CD25 was not a required § 2 district because it was not compact, and thus could not offset the loss of CD23. Id. at 430-31. 432. On remand, the three-judge court’s plan redrew districts 15, 21, 23, 25 and 28. The court drew CD23 to make it an effective Hispanic opportunity district, based on its HCVAP numbers (57.4% HCVAP) and its election performance, and re-united Webb County and placed it in CD28. The court rejected configurations that would place Webb County in CD23 because of the 344a “clockwise” consequences in the map, and found that placing Webb County in CD28 allowed the South Texas districts to be more compact. LULAC v. Perry, 457 F. Supp. 2d 716 (E.D. Tex. 2006); PL-230 (order adopting plan 1438C); PL-231 (opinion); D-648 (map). In this configuration, CD23 contained 17 whole counties, and parts of El Paso County, Bexar County, and Sutton County. 433. In endogenous elections in the redrawn CD23, the Latino-preferred candidate won two of the three elections in 2006, 2008, and 2010. The endogenous election is most important because it tells whether the district is actually performing for minorities. TrA593 (Handley). 434. In the 2006 special November elections under the new map, CD23 incumbent Henry Bonilla received 48.6% of the vote and Ciro Rodriguez (Hispanic, Dem- ocrat) received 19.86% of the vote (there were six Democrat candidates, and Rodriguez received the highest number of votes). D-422 (no obj). In the December 2006 special runoff election, Rodriguez prevailed with 54.28% of the vote, defeating incumbent Bonilla. D- 421; Tr777 (Rodriguez). Rodriguez was re-elected in 2008. Tr777. Rodriguez was the Hispanic candidate of choice. Tr515 (Engstrom). 435. Francisco “Quico” Canseco, a Hispanic Repub- lican, defeated incumbent Rodriguez in 2010. Canseco was not the Hispanic candidate of choice. Tr225, Tr302 (Kousser); Joint Expert Ex. E-2 (Kousser report) at 46; Tr515 (Engstrom) (Rodriguez was the Latino- preferred candidate). Canseco had previously run in 2004 and 2008, but he did not prevail in the Republi- can primaries. TrA565 (Canseco). In 2010, Canseco was the only Hispanic in the Republican primary against several non-Hispanic candidates, and he narrowly won 345a the primary runoff. D-35; PL-1031; TrA566 (Canseco); PL-1030 (showing that he won the primary runoff by 722 votes). In the general election, Canseco received 49.39% of the vote (74,853 votes), and Rodriguez received 44.44% of the vote (67,348 votes). D-35; Joint Expert Ex. E-17 (Alford report) at 4-5. 436. Dr. Engstrom found racially polarized voting in the 2010 general election for CD23. Engstrom Corr. Rebuttal Report (docket no. 307-1) at 25. Incumbent Rodriguez received an estimated 84.7% of the votes cast by Latinos and just 18.1% of those cast by non- Latinos. Dr. Handley also found racially polarized voting in this election. Latino voter turnout was 40.77% of voters, which was low. Tr509, Tr514 (Engstrom); Engstrom Corr. Rebuttal Report (docket no. 307-1) at 25-26. Dr. Alford agreed with Dr. Engstrom’s turnout data. Tr1865. 437. In Plan C100, benchmark CD23 was 62.8% HVAP (30.8% Anglo VAP), 58.4% HCVAP using 2005- 2009 ACS data, and 52/52.6% total/non-suspense SSVR. Benchmark CD23 was a Latino opportunity district. Tr512-15 (Engstrom). Dr. Engstrom found that benchmark CD23 presented a reasonable opportunity for Latinos to elect their candidate of choice; the fact that they did not elect their candidate of choice in 2010 did not mean it no longer provided an opportunity, only that the opportunity was not seized. Engstrom Corr. Rebuttal Report (docket no. 307-1) at 26-27. 438. Benchmark CD23 elected the Latino candidate of choice in only three of the ten elections in the OAG10 (2002 general election for Governor, 2008 general election for US Senate, and 2008 general election for Justice of the Supreme Court). TrA1637 (Downton); TrA303 (Interiano). In Handley’s 2011 Exogenous Election Index, the district performed 40% (2/5) for the 346a minority candidate of choice. Red-400 Report. In her 2014 Exogenous Election Index, the district performed 50% (3/6). TrA595 (Handley); US-687A at 5. In the Task Force’s exogenous election indices, the district performed 7/14 in racially contested general elections between 2002 and 201027 and 3/7 in racially contested general elections between 2006 and 2010. PL-200. 439. Downton testified that there was a disagreement among mapdrawers about whether benchmark CD23 was a § 5 district because of its unreliable performance for Latino candidates of choice. TrA1636, TrA1640. The TLC and Hanna never advised whether it was a performing § 5 district. TrA1637-38 (Downton); TrA1548-49 (Hanna). Hanna never rendered an opinion to the Legislature that 3 out of 10 was affirma- tively a performing district, nor did he even accept that the 10 elections chosen were necessarily the ones to use. TrA1564 (Hanna). It was Downton’s opinion that CD23 was a Latino opportunity district because it was over 50% HCVAP, but it was not a performing district based on the election analysis. TrA1637 (Downton). Downton talked to Interiano, Solomons, and perhaps Hanna about it. TrA1639 (Downton). He did not receive any legal advice that benchmark CD23 was a performing district. TrA1638, TrA1640 (Downton). The OAG did not identify CD23 as one of the performing districts. TrA1639 (Downton). At the end of the day, Downton made the call that it was not a performing district for § 5 purposes. TrA1639 (Downton).

27 The Task Force added four racially contested elections to the OAG10: the 2008 election for Court of Criminal Appeals (Molina won); the 2002 election for Supreme Court (Yanez won); the 2002 election for Court of Criminal Appeals (Molina won); and the 2002 election for Supreme Court Pl. 4 (Mirabal won). PL-200 at 5; PL- 237; PL-238; PL-239; PL-240; PL-241; PL-291; PL-1638. 347a 440. In Plan C100, CD20 was a performing Latino opportunity district in San Antonio represented by Hispanic Democrat Charlie Gonzalez. It was wholly within Bexar County. It was 68% HVAP (21.7% Anglo VAP), 63.8% HCVAP using 2005-2009 ACS data, and 58.1/59.2% total/non-suspense SSVR. It elected the Latino candidate of choice in 10/10 elections in the OAG10. D-141. 441. In November 2010, Eric Opiela sent his “nudge factor” email to Interiano, noting that the nudge factor strategy was “especially valuable in shoring up Canseco.” US-75; PL-1617; Quesada-253; NAACP-651. Opiela also emailed Congressman Lamar Smith (forwarding to Denise Davis, Speaker Straus’s chief of staff), stating that he had been discussing the map with Interiano. He wrote, “I think we should focus on CD23 for enhancement since it is the easiest to reconfigure without the ripple effect – principally by taking more of your district [CD21] in, e.g. your western counties (especially since due to growth, you’re ‘heavy’ in terms of population), and potentially some of CD 20 if we have to enhance Hispanic VAP to offset (problem here is that CD 20 has to grow in size – it’s been growing slower than the state average) . . . there’s a lot more possibilities working on CD 23 first. We definitely don’t have to reconfigure to his detriment since we can play a lot with the large body of people in Bexar to ensure his minority majority status AND make it still Republican – I don’t know yet whether he was the minority community candidate of choice – suspect he wasn’t – but once we get the VTD level election data from Nov. 2 we will know the answer to that question.” Interiano agreed that Opiela was suggesting drawing districts that would appear to be Latino opportunity districts because their demographic benchmarks were above a certain level but would elect a candidate 348a who was not the Hispanic candidate of choice. TrA375 (Interiano). 442. Downton was told that Canseco was not the Latino candidate of choice in 2010, but he did not verify this information. Tr966 (Downton); Downton 8-12-11 depo. at 90; Downton 8-31-11 depo. at 24. Interiano was also aware that Canseco was not the Latino candidate of choice in CD23. TrA309. 443. On April 6, 2011, Opiela emailed Downton a zip file (0406.zip). TrA1679 (Downton); PL-311 Pt. 8 at 434. Downton imported the file into RedAppl as plan hrc1C104. TrA1680 (Downton); PL-587 (plan packet). Plan hrc1C104 incorporated into CD23 some counties north and east of the Pecos River, including Loving, Ward, Winkler, Crane, Upton, Reagan, Schleicher and Crockett Counties, that had been in CD11 in the benchmark. TrA1681 (Downton). All of Sutton County, which had been split between CD11 and CD23, was in CD23. TrA1681 (Downton). Maverick County was not in CD23. TrA1681 (Downton). Using 2005-2009 ACS data, CD23’s HCVAP was 57.9%. TrA1682 (Downton). SSVR for CD23 was 51.9/52.2%. PL-587. Chavez- Thompson received 37.8% of the vote in the 2010 general election for Lt. Governor; Noriega received 47.2% of the vote for U.S. Senate in 2008; Yanez received 50.3% of the vote for Supreme Court Place 8 in 2008; and Molina received 46.4% of the vote for Court of Criminal Appeals Presiding Judge in 2006. PL-587. 444. In advocating for the congressional delegation proposal, Congressman Smith noted that adding a central Texas VRA district “provides us with the legal basis (creating a new Hispanic opportunity district) for not significantly increasing the Spanish Surname Reg- istered Voter percentage in CD 23 which would endan- 349a ger Cong. Canseco, since the only source for those voters would be on the heavily Democratic south side of San Antonio.” US-749; D-611; PL-1025. 445. Downton did a significant amount of drafting for the congressional plan before the special session began. TrA1683 (Downton). By April 6, he had played around some but had not seriously begun drafting a plan. Id. Opiela’s April 6 plan (hrc1C104) was only the fourth plan in his plan account. In May, both the House and the Senate redistricting committees were working on draft congressional maps. Seliger believed that CD23 in the benchmark plan was a Latino opportunity district. TrA219-20 (Seliger). Seliger met with Congressman Canseco during redistricting, and they discussed the configuration of CD23. TrA237 (Seliger). Canseco expressed a goal to be reelected. Id. Seliger’s goal at the start of redistricting was to figure out if there was any way “in political terms” to help Canseco hold the seat. TrA222 (Seliger). The Senate redistricters could not figure a way to help Canseco while keeping the district a Latino opportunity district. TrA223 (Seliger). Doug Davis had VRA concerns with regard to CD23. TrA325-26 (Interiano); PL- 673. Toward the end of the regular session, Seliger and Solomons agreed to move forward with the congressional plan draft that Downton was working on. The configuration for CD23 eventually came from the House and was drawn by Downton. TrA224 (Seliger). Solomons testified that he did not express any goals to his staff during the session with respect to the configuration of CD23. TrA1334 (Solomons). 446. Downton testified that one of his goals in drawing CD23 was to allow Congressman Canseco to have the opportunity to be re-elected as a Hispanic Republican. Tr952, TrA1634 (Downton). Specifically, 350a Downton said he wanted CD23 to perform for Canseco as a Republican district since he was running as a Republican. Downton 8-12-11 depo. at 37. However, Downton made no effort to protect or increase Canseco’s chances of being elected in the Republican primary. Downton did not look at primary elections when drawing CD23 and did not look to see whether Latinos could elect their preferred candidate in the Republican primary. Tr967 (Downton); Downton 8-31- 11 depo. at 27. He also did not have the data regarding the degree to which voters in the Republican primary supported Latino candidates in the areas he was putting in or out of CD23. Tr967-68 (Downton); Downton 8-31-11 depo. at 27. He removed Bexar County precincts that had performed for Canseco in the primary, and he also removed areas that were more favorable for Carillo in the 2010 Republican primary for Railroad Commissioner relative to areas that he kept in CD23. Downton 8-31-11 depo. at 38-39; Tr968-70 (Downton). Downton thought Canseco could be nominated in a racially contested Republican primary based on information from Canseco’s office and Canseco’s satisfaction with the district, but no other information. Downton 8-31-11 depo. at 35-36, 47. 447. Downton and Interiano said the CD23 goals were to give Canseco his best chance to be reelected, while maintaining or increasing the Hispanic percentages (SSVR, HCVAP, HVAP), which they felt needed to be done to comply with the VRA. Tr952, TrA1634- 35 (Downton); Downton 8- 12-11 depo. at 73-75, 96; TrA376, Tr1454-55 (Interiano); Interiano depo. vol 3 at 70. Downton said these goals were separate but were not mutually exclusive. TrA1635 (Downton); Downton 8-12-11 depo. at 110. He noted that HCVAP was their “principal benchmark,” that a district becomes a Hispanic opportunity district for § 2 purposes when 351a the HCVAP crosses the 50% threshold, that CD23 was already across that threshold, and they wanted to at least maintain its current level. TrA1635 (Downton). Downton believed that if a district was majority- HCVAP, then by definition it was an opportunity district in compliance with the VRA. Downton 8-12-11 depo. at 110-112. He acknowledged that whether any particular person can get re-elected depends on the person, their history with a particular political party, their roots in the community, and “lots of factors.” Downton 8-12-11 depo. at 110. 448. Downton said that, to give Canseco his best chance, he needed to increase the number of Hispanic Republicans. Downton 8-12-11 depo. at 75-76. When asked how he identified Hispanic Republicans in order to put them in CD23, Downton said he looked at SSVR and performance of specific precincts in deciding whether to move them in our out. Tr955; Downton 8- 12-11 depo. at 79. He could see what percentage of a particular precinct voted for John McCain and what percentage of a particular precinct was Hispanic, and if there were two precincts that were both 60% His- panic, he would pick the one that had a higher percentage of people voting for McCain. Tr956 (Downton); Downton 8-12-11 depo. at 76-77. More precisely, if a majority Hispanic area tended to have more votes in those precincts for the Republican candidate, he would be more likely to include those precincts. Tr965 (Downton). Downton acknowledged that a precinct could have a higher number of votes for McCain because McCain was the preferred candidate of Anglo voters and they were turning out at a higher rate than Hispanic voters in the precinct. Tr956; Downton 8-12- 11 depo. at 77. Downton said they considered that when mapping, but did not have any other way to make determinations because “precinct results are the 352a smallest voting block [data] you can get.” Downton 8- 12-11 depo. at 77. Downton also acknowledged that a precinct could have a higher vote count for McCain because there was a greater degree of racial polarization in candidate preference in that precinct (i.e., Hispanics are voting the same in each precinct for the Latino-preferred candidate, but Anglos are more polarized against the Hispanic candidate and thus more Anglos are voting for the other candidate). Tr957; Downton 8-12-11 depo. at 77-78. He stated that he did not think they considered that, but it would not have made a difference anyway because this was all the data that they had to go on. Downton 8-12-11 depo. at 79. Downton did not display any running statistics in RedAppl for political races that he thought of as indicative of whether a district could elect a Latino- preferred candidate, but he looked at the OAG 10 reports that indicated whether a Latino-preferred candidate could be elected. Tr954-55, 958-59 (Downton). 449. Canseco said that his goal in redistricting was to maintain the integrity of CD23 as a Hispanic- majority district because that is the area of the state he identifies with and an Anglo-majority district would not be a good fit for him. TrA569-70. He also wanted a “fair shake” as a Republican. TrA578 (Canseco). At one point Canseco rejected acquiring a purely Republican area as not a good fit and because he “just wouldn’t make it through a primary.” TrA570-71 (Canseco). He wanted to maintain the character of the district; he did not want a “Farenthold district” by going northwest, and wanted to maintain “the border integrity, the rural integrity, the San Antonio aspect” of CD23. TrA570, TrA579 (Canseco). Canseco said the constraints “were getting rid of 150,000 or more in population, preserving the characteristics of the 23rd District as an opportunity district, a minority opportunity 353a district, and making sure that it works, so that it is competitive within the [ambit of the] Voting Rights Act.” TrA572 (Canseco). Canseco also requested that they draw his house into CD23, and they accommo- dated him. Tr952 (Downton); Downton 8-12-11 depo. at 74; Downton 8-31-11 depo. at 34 (they moved in “some territory in sort of the north side of San Antonio to make sure Mr. Canseco lived in his district”). 450. In Downton’s plan hrc1C130 (PL-1134), “Con- gressional Ryan Merge 5/20” created May 20, the configuration of CD23 included the same counties as in the April 6 map sent by Opiela (hrc1C104). Specifi- cally, the counties north and east of the Pecos River were in CD23 and Maverick County was not. TrA1683- 84 (Downton). The CD23 boundaries in El Paso and Bexar County are different from Opiela’s plan. The 2005-2009 ACS HCVAP of CD23 is 58.2%, close to the 57.9% in Opiela’s plan. TrA1684 (Downton).28 The reconstituted election results for Chavez-Thompson, Noriega, Yanez, and Molina were all within .2% of Opiela’s plan. PL-1134. 451. In Downton’s plan hrc1C132 (PL-1136), “Con- gressional Ryan Merge 5/25” created May 25, CD23 included the same counties. TrA1687-88 (Downton). Its HCVAP is 58.2%. There are additional drafts between May 20 and May 30, and the versions of CD23 in these maps were identical outside of Bexar County and El Paso. TrA1688 (Downton). In plans hrc1C134 and hrc1C135, the HCVAP of CD23 is 59.2%. PL-1138;

28 Interiano’s RedAppl account also has a May 20 plan (strjC106) called “Congressional Proposal Map 2.” TrA328 (Interiano); PL- 1338. At trial, Interiano was asked about the SSVR and election performance of CD23 in this plan, which was better than the enacted plan. TrA329-30. However, many of the districts in this plan, including CD23, are not at ideal population. 354a PL-1139. Although the HCVAP increases from 58.2% to 59.2% between hrc1C130/hrc1C132 and hrc1C134/ hrc1C135, the performance of Chavez-Thompson, Noriega, Yanez, and Molina increases by only .1 or .2%. 452. On May 28, Interiano asked Downton, “Any guidance on your 23? Have you been able to make any of the changes that we all discussed?” Downton replied, “Have it over 59% HCVAP, but still at 1/10. There has to be some level of HCVAP where it doesn’t make a difference what the election results are. It is more Hispanic than the other two San Antonio based districts.” US-630 (the hearsay objection to this exhibit is overruled because it is not admitted for its truth); PL-1659. 453. On May 30, Lamar Smith emailed Opiela asking for Downton’s email and asking if it would help Canseco “if I gave him 3k more in bexar (either gop or hispanics) and took edwards co in exchange.” D-612. Opiela responded, “I don’t think we mess with quico’s district–for your sake and his. His is barely performing (or not depending on your measure) right now; add Rs (which will be Anglos) and you put a neon sign on it telling the court to redraw it. Bring down your numbers and you’ll have a Demo opponent every time. And they won’t be Lainey Melnick.” D-612; US-629; Quesada-144. Smith responded, “Still want to make offer re edwards. Only 3k. Maybe .1 percent but cld help quico.” D-613. 454. Plan C125 was released May 31. It contained the new majority-HCVAP district CD35 connecting Travis and Bexar Counties. The new CD35 was 52.8% HCVAP using 2005-2009 ACS data and 44.8/46% total/non-suspense SSVR. CD23 was 59.2% HCVAP and 52.8/53.2% total/non-suspense SSVR. CD20 was 355a 57% HCVAP and 49/49.5% total/non-suspense SSVR. D-548.6. 455. In Plan C125, the border between CD16 and CD23 in El Paso was very irregular, with an extension of CD23 into CD16 in El Paso County. The irregular boundary reflected an attempt to include in CD23 portions of El Paso County that would increase Canseco’s election chances. Tr964 (Downton); Downton 8-12-11 depo. at 97- 98.29 That boundary remained the same until Solomons’ West Texas Amendment Plan C170. Downton 8-12-11 depo. at 98-99.30

29 At the August 2014 trial, Downton testified for the first time that he believed “that version of El Paso came from the Republi- can Congressional Delegation map.” TrA1640-41. However, this testimony is inconsistent with Downton’s testimony indicating that he drew the map for CD23, including in El Paso. Further, the map boundary lines do not match with the congressional delegation map proposal (PL-311 Pt. 2 at 24) or the map sent by Opiela on April 6, hrc1C104 (D-510). Thus, Downton’s 2014 testimony is not credible. 30 Although the CD16/CD23 boundary was not changed until late in the process, Downton testified that the change originated earlier based on concerns raised by Rep. Pickett and Congress- man Reyes (the CD16 incumbent). Downton stated that at some point after Plan C125 was released or Plan C141 came out of the Senate, Rep. Pickett called Downton and asked why the El Paso map he and Congressman Reyes had worked on was not included in the map. Downton 8-12-11 depo. at 99. Downton testified that he told him he had not been aware of the map, and that he then worked with Pickett’s staff to prepare a committee amendment. Downton said he thought Pickett would present that amendment at the June 9 committee hearing, but Pickett told him that Congressman Reyes had changed his mind, and he did not offer the amendment. Downton testified that, after discussing the matter with Solomons, they decided to change the border anyway and used Pickett’s plan to avoid any claims of gerrymandering and to have a cleaner line because Reyes and Pickett had raised concerns about the shape. TrA1644, TrA1649, Downton 8-12-11 356a 456. Maverick County, which had been wholly within CD23, was wholly in CD28. TrA1641 (Downton). Zavala and Dimmit Counties were also moved from CD23 to CD28. Nearby Frio and La Salle Counties were moved from CD28 to CD23. To the north, several counties (Loving, Winkler, Ward, Crane, Upton, Regan, Schleicher, and part of Sutton County) with about 33,000 people were moved from CD11 to CD23 (as in the prior drafts and in Opiela’s proposal). Tr962 (Downton). Downton was not able to explain why these specific counties were included in CD23. Tr963, TrA1681. 457. At the time of redistricting, benchmark CD16 was overpopulated by about 59,000, benchmark CD23 was overpopulated by 149,000, and benchmark CD20 was overpopulated by 13,000. Downton acknowledged that he could have just shrunk back CD23 to reach ideal population. Tr962. 458. The new Latino district CD35 came into Bexar County and San Antonio and took some of the population, including some of downtown San Antonio, that had been in CD20. Downton testified that because they were creating CD35 as a § 2 district, it had to be over 50% HCVAP, but they also had to maintain the existing majority-Hispanic status of the surrounding districts. TrA1642. He stated, “So in creating District 35, originally, we took a lot of Hispanics out of District 20 to put them into District 35.” TrA1642-43. That depo. at 38-39, 99-100. However, the El Paso configuration in Plan C170 and the final enacted plan do not look like the plan labeled as “Pickett El Paso?” (hrc1C141), “Potential El Paso” (hrc1C142), “El Paso Floor Amendment” (hrc1C178), or “El Paso Floor Amendment 2” (hrc1C179) in Downton’s RedAppl account. Thus, the testimony that the CD16/CD23 border in El Paso came from Rep. Pickett is not supported by the evidence and is not credible. 357a reduced the HCVAP of CD20 below benchmark (from 63.8% to 57%), though it was still above 50%. TrA1643 (Downton). The new CD35 also took some of the south side of San Antonio that had been in CD23. 459. Hanna was familiar with CD23 from the prior redistricting litigation and had § 2 and § 5 concerns about the district and its performance. TrA1515-16 (Hanna). At Hanna’s request (based on his own concerns and possibly also based on concerns raised by Doug Davis), Archer did some analyses on CD23 as proposed in Plans C125 and C130. TrA644-46 (Archer); TrA1515-17 (Hanna). After doing some independent analysis, Archer had some concerns about whether CD23 was a Latino opportunity district. TrA644-45 (Archer); US-609 (analysis). His analysis was done fairly quickly. TrA647 (Archer). He pulled all the election returns for 2006, 2008, 2010 general elections for the existing district and printed them; these reports are available to everybody who has RedAppl. TrA647 (Archer). Archer looked at certain, primarily statewide, elections to try to identify the Hispanic candidate of choice and then to see how they performed in the proposed district. TrA648 (Archer). Archer compared the SSVR and election returns of CD23 and CD15, which he considered a “solid Hispanic opportunity district,” and looked at racial bloc voting analyses on CD23. TrA646, TrA655 (Archer). The bloc voting analysis was done through software that TLC had developed in 2000, but it was put “on a back shelf” and only Archer had access to it. TrA649 (Archer). The scatterplots indicated “a pretty high likelihood [of] polarized voting.” TrA651-52 (Archer). With regard to election results and SSVR, Archer made notes about the SSVR in general elections in CD23 (2002- 53.2%, 2004 - 54%, 2006 - 53.9%, 2008 - 53.4%, and 2010 - 358a 52.8%). US-609 at 13 (the hearsay objection is overruled because this is not admitted for its truth). But Archer did not draw a conclusion concerning whether the numbers were consistent with Hispanic electoral opportunity. TrA654 (Archer). He testified that the numbers “make you think at least there is a theoretical competitiveness to that district for Hispanic voters, but you immediately think of, is the polarization of white voters so great that they might outvote a slightly less [cohesive] bloc vote of Hispanics? Is the turnout different? Is the data crummy for the – or at least suspicious? Is it within the margin of error?” TrA654. Archer noted that Chavez-Thompson, the Latino candidate of choice, was “trounced” by Dewhurst in CD23 in 2010 in the Lt. Governor’s race, and that if he saw that “happening consistently, [he] would be concerned that you don’t have the right electoral activity, the right electoral dynamics for Hispanic-preferred candidates to have an opportunity to be elected.” TrA658. He felt that if 2008 were a representative year, this was a competitive district and possibly “a perfectly good, legal district,” but if 2010 were more representative, the district had problems. TrA661. Archer wanted to pass the information along but did not draw a conclusion about CD23. TrA675 (Archer). Archer noted that the other lawyers involved might be relying on other data, that SSVR “all by itself in a vacuum” could theoretically be used to argue that this is a competitive district that gives Latino voters a reasonable opportunity to elect candidates of their choice, and “[i]t depends on what law you apply to that” such that “Doug and the other attorneys very well understand that there are a lot of different ways of measuring opportunity or success and are going to have to draw their own conclusions.” TrA676. Archer said he did not have time to look at all the data and factors, including 359a HCVAP. TrA668-69. He expressed his concern to Hanna. TrA661 (Archer); TrA1517 (Hanna). He passed on his summary about 2008 and 2010, essentially noting the SSVR and election returns, and said, “you need to look at this closely and decide for yourself what you have got here, because 2010 could indicate a problem.” TrA662 (Archer). He did not discuss his thoughts directly with Downton, but he was confident that Hanna passed them on. Id. Hanna shared this concern with Doug Davis and Interiano, but could not recall if he told Downton. TrA1518 (Hanna); PL-1622. Archer said he did not really track the issue after that. TrA662-63. 460. At the June 2 HRC hearing on Plan C125, Rep. Menendez (Hispanic, Democrat) spoke on behalf of himself and Congressman Charlie Gonzalez (CD20), stating that a minority opportunity district should not be based strictly on 50% numbers, but on whether it is effective. D-601 at 23-24. He continued, “So given that premise, I would like to speak briefly about the 20th Congressional District. Foremost, it was the first district in the State of Texas to elect an Hispanic to Con- gress in 1961. It has always been the seat that remains entirely within Bexar County, and it also represents the heart of the people of San Antonio physically and culturally. And we would like and feel that we deserve to have a district that solely represents their interest in our community. The proposed [HB 4] does not do that. Furthermore, it has also been largely an urban district that has been compact. And, finally, it has always been a majority Latino district that has effectively elected the candidate of their choice.” Id. at 24. He further stated, “The 2010 census also indicates the 20th Congressional District needs to lose about 13,217 people in order to achieve the ideal population of 698,488. Because of this our community asserts that 360a additional congressional districts can be created around the 20th Congressional District with minimal impact to the current lines. The proposed HB 4 map changes the district in very significant ways. First of all, it decreases the percentage of registered His- panic voters below 50 percent [to 49/49.5% total/non- suspense SSVR]. It takes out for the first time in its history, it takes out the cultural soul of the district by removing large parts of Edgewood. Edgewood being the school district that started Edgewood versus the State of Texas, and the West Side, which are core communities of interest of the current 20th Congres- sional District, which I’ve also represented in this State House for the last ten years. And it puts [it in CD35] and it stretches that district all the way to South Austin . . . . I respectfully ask you and this committee to maintain the integrity of the 20th Con- gressional District as much as possible when creating additional districts and making changes to the existing districts in Bexar County.” Id. at 25-26. When asked by Rep. Hilderbran what percentage SSVR was necessary for an effective district, Menendez said, “In order for effectiveness on election day I personally believe it needs to be in the 56 percent, not 51, not 52. But I think it needs to be 55 plus.” Id. at 28. Solomons noted that “District 20 remains a Hispanic majority district in Bexar County. Its Hispanic – Hispanic citizen voting age population was 63.8 percent. . . . And the new map dropped it to 57.0 percent. But that change was actually necessary to create a new His- panic majority district, District 35. And our – our folks who looked at all that, but from the legality side and retrogression and so forth, don’t – do not believe it’s retrogressive. And a retrogression analysis shows that the Hispanic community still elects a candidate of its choice in nine out of ten elections. So even though it 361a did drop, we created a new Hispanic majority district and it’s – and they could still elect – still elects a candidate of its choice in a retrogression analysis.” Id. at 32. Menendez said, “I definitely appreciate the fact that you have looked at the issue of retrogression. I’m glad that the committee is looking at it seriously, and I look forward to working with y’all.” Id. at 33. 461. Plan C130, Seliger’s statewide substitute plan to C125, was made public on June 2. Between Plan C125 and C130, some population was shifted between CD20 and CD35 in San Antonio. CD35 took more of downtown San Antonio from CD20. The HCVAP of CD20 was increased to 57.8% and SSVR was increased slightly to 50/50.5%. The HCVAP of CD35 dropped from 52.8% to 52% and SSVR was decreased slightly to 43.7/44.8%. D-549; D-554. 462. Opiela emailed Interiano on June 2 with the subject “cd20” and asked, “Is the regression analysis back on C130? does it perform 10 of 10?” D-616. Interiano responded, “Don’t know if Ryan requested it from the AG. And it would take several days for the AG to do it. Either way, I’ll put in the request now.” Opiela was concerned throughout the process that CD20’s election performance not retrogress from the benchmark (which was 10/10). 463. At the June 3 SRC hearing, when asked about whether CD23 was still effective, Seliger responded that it had a total SSVR of 52.8% (in Plan C125) and “that reflects a – the, the Hispanic population in Dis- trict 23.” D-602 at 24. Sen. Rodriguez complained that the map did not have two new minority districts and that it put half of El Paso County into CD23 with San Antonio. He complained that CD23 was drawn to protect Canseco rather than reflecting communities of interest and geography. Id. at 41-42. Rep. Menendez 362a testified on behalf of Congressman Gonzalez regarding CD20. He complained that the SSVR was decreased from an effective percentage to 50% and that the map took out downtown and large parts of Edgewood ISD and core communities of interest in benchmark CD20. Id. at 44. Nina Perales complained that Plan C130 unnecessarily reduced Latino electoral opportunity in CD20 and CD23 and cut up San Antonio neighborhoods. She said, “Plan C130 reduces the Spanish Surname Voter Registration in District 20 to 50 percent. That is a dramatic drop of 8 percent. As a result, the votes garnered by Latino candidates are also dramatically reduced in the new version of the district. This reduction is unnecessary and makes it more difficult for Latinos to elect their preferred candidate in District 20.” Id. at 78. She continued, “Although Plan C130 does not reduce the percent Spanish surname registered voters in Congressional District 23, it nonetheless retrogresses that District by altering its geography and voter composition. Plan C130 retrogresses CD23 by removing large portions of the South Side of San Antonio, as well as Maverick, Zavala and Dimmit Counties, and adding in Frio, La Salle, Loving, Winkler, Ward, Crane, Upton, Reagan, Schleicher, and a portion of Sutton County. As a result, Latino voters in CD23 no longer have an opportunity to elect their candidate of choice. The changes in South Texas are gratuitous and suggest an intent to obstruct, or wholly prevent Latino voters from electing their preferred candidates in Districts 20, 23, and 27.” In laying out Plan C130, Seliger said they had raised some of the matrix percentages for CD 20 (which he explained are things like HCVAP) in response to prior testimony from Rep. Menendez. Id. at 180. Plan C130 was adopted, and Plan C136 was the plan as passed out of the SRC. 363a 464. A racially polarized voting analysis from the OAG showed that in Plan C136, Hispanic performance in CD20 was 9/10, compared to 10/10 for Plan C100. D-141_0004. 465. On June 4, Denise Davis sent an email to Interiano with the subject “Re: CD20 Gonzales.” PL- 1623; US-182 at 6; US-621; US-624. Davis asked, “Can you tell me how the tweak for that’s going?” Interiano responded, “Not sure that there is going to be a tweak. [Opiela] had told me that the RNC was working on something but that they couldn’t get it to work. We’ve been talking to the AG daily and have asked them to review CD20 again but they don’t appear to be concerned. I know that [Opiela] is. Hope that helps but feel free to call me if you want to discuss further.” Davis replied, “Lamar called Speaker and was worried. What do Phillips and Cooper think?” Interiano answered, “Last I talked to [Tom] Phillips [an attorney at Baker Botts] he was on with it. The reason why the numbers drop in CD20 is because a new district is created. But it is still a performing district which the AG has reviewed. But why don’t I see if he can come in on Monday and we can discuss it in more detail.” Davis then wrote, “Ok.” On June 6, Interiano emailed Tom Phillips and Radney Wood to see if they could meet to discuss the congressional map. PL-1670. Interiano then asked Downton and Hanna if they could meet with them “to go over the issues that are continuing to be raised re CDs 20, 23, and 29?” Id. 466. During the June 6 Senate floor debate, Sen. Rodriguez complained that the CD16 and CD23 border in El Paso split communities of interest to protect Canseco and was a potential violation of the VRA. D- 22 at A-10. Sen. West proposed statewide substitute Plan C121, the “Fair Texas Plan.” West stated that 364a existing opportunity districts 23, 27, and 25 were preserved in Plan C121. Id. at A-18. It was tabled. The plan as passed out of the Senate was Plan C141. 467. On June 6, Opiela sent Downton an email and proposed map, stating “Here’s a change of bexar county dists which will get CD20 to 56 ssvr while keeping CD35 at 44 ssvr.” PL-311 Pt.8 at 445; D-618. It had an attachment called C136-CD20Mav.zip. On June 7, Downton created plan hrc1C162 in his RedAppl account with the comment “San Antonio-Maverick.” D-508.1. It included a CD20 that joined parts of San Antonio/Bexar County with Maverick County (and included part of Atascosa and all of Frio, Zavala, Dimmit Counties). PL-793. The HCVAP of CD20 was 62.3% and SSVR was 56/56.9%. This same configuration of CD20 was in hrc1C163 with the comment “Final Committee Sub” on June 7. PL-794. However, later maps on June 7, including hrc1C165 with the comment “Final Committee Sub 2” do not include this version of CD20. PL-796. Interiano testified that the mapdrawers wanted CD20 to remain within Bexar County, and when the delegation proposed a CD20 that went outside Bexar County, “the House” did not accept it. TrA363. 468. Late afternoon on June 7 (5:22pm), Opiela sent an email to Downton and Interiano titled “20 at benchmark” with the attachment “Odessa in 23.zip.” D-620; PL-311 Pt. 8 at 448. The email stated, “run this and see if it is 10 out of 10.” D-620; TrA356 (Interiano). Downton said Opiela “did not like the committee’s configuration of District 20” and in this plan Opiela “suggested taking the City of Odessa, which is in Dis- trict 11, and moving it into District 23 because there was a Hispanic population there.” TrA1645. Downton testified, “For Section 5 reasons, [Opiela] thought it 365a needed to continue to perform at ten out of ten elections. And so he was sending us a map for us to check if his particular suggested configuration had District 20 performing at ten out of ten elections.” TrA1645-46. This proposal was not accepted. TrA1647 (Downton). Although this was not the only map in Downton’s account that included Odessa in CD23 (plan hrc1C196 dated June 13 also did), Downton said it was never really an option for Odessa to be taken out of CD11 and put in CD23 because the congressman of CD11 (Conaway) would not have agreed. TrA1647. 469. Opiela was particularly concerned with CD20, and he wanted the HCVAP to be at the benchmark level. TrA1646 (Downton). Downton testified that his view and Solomons’ view was that CD20 was a His- panic opportunity district because it was over 50% HCVAP. TrA1646. However, for § 5 purposes, the performance of CD20 had decreased and they did not think it was a bad idea to try to raise it back up. TrA1646 (Downton). 470. On the morning of June 8, Solomons released his statewide substitute Plan C144. D-506; PL-1122. Plan C144 made a number of changes to CD20, CD21, CD23, CD28, and CD35 in Bexar County. Also that morning, Opiela emailed Downton and Interiano on behalf of Lamar Smith requesting that a specific Bexar County precinct be moved from CD35 to his district CD21. PL-311 Pt. 8 at 452; D-621. Opiela explained “it’s in CD35 in Plan C136. It’s the SA Country Club adjacent to SJS’s house. Giving it to lamar should help improve ssvr a fraction in CD35. You will also have to give pct 4101 to CD21 as well because it would be orphaned (357 people).” D-621. Downton responded, “I’ll work on it” and later “Change has been made.” PL- 311 Pt. 8 at 452; D-621. 366a 471. On June 8, Downton created and saved plan hrc1C180 with the description “Eric El Paso.” D-508.1; D-523 (map). The comments say “copy of plan hrc1C129” which was an earlier map described as “Eric 5/20.” D- 508.1; D-511 (map). Both maps have the same configuration of the CD16/CD23 border in El Paso. The maps were prepared by Opiela, but they were not adopted. TrA1648-49. 472. Later on June 8, Downton emailed Interiano an OAG RPVA summary for districts 20, 23, and 35 in a proposed plan (XOAGC118) and for CD23 in a proposed El Paso floor amendment. PL-1665; US-626; Quesada-241. The RPVA shows CD20 performing at 9/10 in the proposed plan XOAGC118. It also shows CD23 at 3/10 in the benchmark and 1/10 in XOAGC118 and shows slightly decreased Republican performance in CD23 in the El Paso floor amendment compared to proposed plan XOAGC118. 473. At the June 9 HRC hearing, Solomons offered his statewide substitute Plan C144, and it was adopted and incorporated into the House Committee Report Plan C149. As in Plan C125, Plan C149 has the same El Paso configuration with protrusions, and Maverick County is still wholly in CD28. TrA1335; D-559.1. However, several changes were made to the districts in Bexar County. CD35 took more of downtown San Antonio and parts of central San Antonio from CD20, and CD20 moved to encompass portions of the south side of San Antonio inside Loop 410 that had been in CD35, and then stretched almost completely across Bexar County to the southeast. CD20’s HCVAP increased from 57.8% to 60%, and its SSVR increased from 50/50.5% to 52.3/52.8%. CD23’s HCVAP dropped slightly (from 59.2 to 58.9%) and its SSVR dropped by .2% (from 52.8/53.2% to 52.6/53%). CD35’s HCVAP 367a dropped from 52% to 50.2% and its SSVR dropped from 43.7/44.8% to 41.7/42.8%. TrA1647 (Downton); TrA1336 (Solomons). 474. On June 9, Downton created plan hrc1C185 in his RedAppl account. D-508.1. The comments said, “Maverick switch” and it included a CD35 that joined Travis County with Maverick County. PL-816. Not all districts in the map were at ideal population. On the morning of June 10, Lee Padilla (NRCC) emailed Interiano saying, “There is no way in hell they go for TX-35 snaking all the way from Austin to Maverick County.” Quesada-68 at 50. Interiano responded, “I already told that to Eric [Opiela]. Is he telling you guys to draw that?” Padilla replied, “No brother, it was a brainstorm of what hasn’t been done, because honestly, everything else has. I know we disagree on the 20 thing, but as I look at the #% snake to Maverick, if you guys won’t take 20 going to Maverick, why the hell would take 35 doing it? Just venting, this whole process has been frustrating but it is what it is . . . . Eric [Opiela] speaks for us, not Dale, Tom, Dub or any of those . . . .” Quesada-68 at 50. 475. On the afternoon of June 10, Opiela forwarded an email with an El Paso map attachment (El Paso_June 10.pdf) from the NRCC to Downton and Interiano. PL-311 Pt.8 at 456-457; D-623; D-638 (attachment); Quesada-68 at 30. The subject of the email was “does this look better for El Paso?” Opiela wrote, “Still waiting for block file. Will forward when I get it.” Quesada- 68 at 30. This map maintained the northern CD23 extension into CD16 but got rid of the southern extension. D-634. Downton responded, “How are the Hispanic numbers compared to benchmark?” At 7:12 p.m. Opiela sent an email to Downton and Interiano with the subject “C149-El Paso Amend Final” and the 368a attachment “C149-ElPaso.zip.” D-624; Quesada-68 at 10. It stated, “Here is the block file for the El Paso amendment 51.73 McCain 52.99 SSVR compared to C149 which was 51.94 McCain and 52.57 SSVR.” TrA357 (Interiano). Downton testified that he had told Opiela that Pickett had raised issues with El Paso “looking ugly” and that changes might be coming, and that Opiela was advocating for his preferred changes. TrA1650, TrA1654. Opiela also had concerns about CD20 not being at the benchmark, so he sent alterna- tives that he liked better. TrA1653-54 (Downton). 476. At 9:59 p.m., Opiela sent Interiano and Downton an email entitled “Couple of options” that attached “two plans that should (1) improve CD 23’s hispanic performance while maintaining it as a Republican district and (2) in the case of one get CD 20 to benchmark, and the other get it within 1 point of benchmark on SSVR.” PL-311 Pt. 8 at 459; D-625; Quesada-68 at 6. The two attachments are C149-CD23modEctor. zip and C149-23mod.zip. D-635; D-636. In Opiela’s proposal C149-CD23modEctor, part of Ector County, including Odessa was placed into CD23, as well as all of Maverick County. D-635. Opiela’s proposal C149- CD23mod did not include any of Ector County in CD23 and changed CD20, CD21, and CD23 in Bexar County as compared to C149CD23modEctor. The configuration of Bexar County and El Paso County in these maps is different from Solomons’ later amendment Plan C170. 477. Opiela asked Interiano to “have them run against election performance.” Interiano responded at 10:10 p.m., “Only the AG and the Legislative Council can run those reports. We’ll have them run on Monday but there are several other issues we are trying to address.” PL-311 Pt.7; D-625; Quesada-68 at 8. The 369a next morning, Interiano wrote Downton, “I can’t upload this at home . . . but think this may actually be a good option since it keeps CD20 entirely within Bexar County. Curious to see what it does to CD35. Let’s discuss it tomorrow.” PL-311 Pt. 8 at 459. Downton wrote back, “Sounds good.” Id. 478. On June 11 at 7:54 a.m., Opiela sent Interiano an email with the subject “SA and cd 23 reconfigure PDF” and the attachment “CD23-C149m.pdf.” D-626; PL-1621; Quesada-68 at 37-39; D-639. The email contains a forwarded email from Opiela to Lamar Smith stating, “I was able to keep Quico substantially the same GOP and get Charlie [Gonzalez of CD20] to 57.7 SSVR. I replaced the entire south side of SA inside loop 410 with Maverick and Zavala [both Maverick County and Zavala County were in CD28 in Plan C149]. I can get it the extra 1 SSVR point if I took W Odessa, but I know we’re not going there with Conaway. Wanted to at least get it out there as an option. I think it will probably increase Quico’s and Charlie’s Hispanic Performance one election. I think we should at least discuss it.” D-626. At 9:34 a.m., Interiano responded, “Did you make any changes to CD35 where it would have dropped the CVAP below 50%?” D-627. Interiano testified that mapdrawers felt that CD35 needed to stay above 50% CVAP or SSVR, but the Republican congressional delegation were not convinced it needed to be above 50%. TrA363. 479. On Monday June 13, 2011 at 7:23 a.m., Opiela sent an email to Interiano (copying Downton) with the subject “optimized Friday’s Plan” and an attachment “C149-CD23modoptoep-nomav.zip.” TrA364 (Interiano); TrA1690 (Downton); PL-311 Pt. 8 at 461; US-195 at 306; D-628; D-637 (attached map). In this plan, Zavala County, Dimmit County, half of Maverick County, and 370a part of Atascosa County are added to CD23. It also contains a new El Paso configuration from Opiela’s prior maps. Counties north of the Pecos River—Ward, Crane, Upton, Reagan, Crockett, Schleicher, and Sutton—are in CD23. D-637; TrA1657 (Downton). Opiela wrote, “This is the best I could do. Only 17,360 in Maverick for CD 23. 52.4% McCain. All other districts at benchmark. Probably could get rid of all of Maverick if did Ector, but right now only districts changed are 16, 23, 28, 20, 35, 15. 20 doesn’t leave bexar county, and 35 doesn’t leave Bexar in the south either. Quico [Canseco] has not yet seen/signed off on this, but could you please run the election performance on this one (in lieu of the non-Ector one I sent Friday). Let’s discuss this morning.” D-628. Interiano testified that this was Opiela’s proposal trying to meet the confines of the parameters “we” had set, specifically, “meeting the benchmarks and also having 20 not leaving Bexar County.” TrA364. Downton testified that this plan gave him the idea to split Maverick County. At trial, he testified that Congressman Canseco had asked them not to put any of Maverick County in CD23, and they were trying to respect that request. Downton said that “Maverick County does not have a particularly good record of voting in high numbers for Republicans.” Tr963; Downton 8-12-11 depo. at 87. At his deposition, Downton said they initially put Maverick County into CD28 because “it fit fine there” along the border with Webb, Zapata, and Starr Coun- ties, and if Maverick County would make CD23 less Republican, then it was less likely they would include it in CD23. Downton 8-12-11 depo. at 89-90. Downton said they needed more Hispanic population in CD23 to offset changes in Bexar County, and since Opiela worked for Canseco, this was “the go ahead that we could look at putting part of Maverick into 23” even 371a though Canseco had not signed off on it. TrA1657- 1658, TrA1691; Downton 8-12-11 depo. at 35, 86-87. They chose to split Maverick County because putting all of Maverick County into CD23 would have increased Hispanic voting strength in CD23 and hurt Canseco’s ability to be reelected. TrA1694 (Downton). 480. At 8:00 a.m., Interiano asked Opiela when he could come down to the redistricting office. PL-311 Pt. 5; D-629; US-195 at 316, TrA313 (Interiano). Opiela responded at 8:04 a.m., “45 mins.” PL-311 Pt. 6. Interiano responded at 8:08 a.m., “Sounds good . . . I’m uploading it right now . . . see you here.” D-629. At 8:16 a.m. June 13, Interiano emailed Hanna and asked him to go to the redistricting office. TrA1525 (Hanna); US-748. Hanna responded, “Around 9 OK?” US-748. As noted, the redistricting office was a shared office between the Speaker and the redistricting committee, and it was Interiano’s and Downton’s primary office. TrA315. Although both Opiela and Hanna responded that they would come to the office, no witnesses recalled the meeting. TrA317 (Interiano); TrA1658, TrA1690 (Downton); TrA1525-1527 (Hanna). 481. Interiano uploaded Opiela’s plan into RedAppl around 8 a.m. The plan became strjC116 in Interiano’s RedAppl account, with the description “New South Texas Amendment,” created at 8:15 a.m. and last modified at 8:32 a.m. TrA313-14; 366; US-664; D-539.1. At 8:34 a.m., Interiano received by email a Red 106 report on plan strjC116 from the TLC. TrA314; US-754. CD23 SSVR was 52.8/53.6%. 31 PL-1348; D-541.6. HCVAP was 56.8%. Chavez-Thompson received 36.9% of the

31 At trial, Interiano testified that non-suspense SSVR was 54.1%, but this testimony does not match the exhibit being discussed. TrA332, PL-948/PL-1348. 372a votes in CD23 for Lt. Governor in 2010; Yanez received 50.0% of the votes in CD23 for Supreme Court Place 8; Noriega received 46.9% of the votes for U.S. Senate in CD23; and Molina received 45.7% of the votes for Court of Criminal Appeals presiding judge. TrA332- 33; PL-948/1348; D-541; US-739. Exhibit D-674 com- pares strjC116 with Solomons’ later floor amendment Plan C170. Although both plans split Atascosa County, the boundary line differs in some areas. The Bexar County boundaries differ as well, as do the El Paso boundaries, and a small portion of the Maverick County split. 482. At 8:22 a.m. Downton re-sent Interiano the RPVA Summary for XOAGC118. TrA527 (Flores); PL- 1665. 483. Interiano’s account also includes plan strjC117 titled “Floor Amendment” with the comment “Slight Revision,” created and last modified June 13 at 12:45 p.m. D-539.1. Like Opiela’s plan, it contains half of Maverick County, part of Atascosa County, and all of Zavala and Dimmit Counties in CD23. CD23 contains only part of La Salle County (Plan C149 and Opiela’s plan had all of La Salle County in CD23). It differs from strjC116 in Atascosa County, Bexar County, El Paso County, and Maverick County, and contains a version of El Paso that is very similar to the final map. D-542.2. Bexar County and Maverick County look the same as the final map. D-542.2; PL-949/1349; US- 731HHH. 484. Downton’s RedAppl plan list contains plan hrc1C187 (US-731NN, PL-818; D-535) labeled as “Poss- ible Floor Map” that was created at 8:33 a.m. June 13. US-658; D-508.1, PL-549. This plan was last modified 373a at 2:27 p.m. that day. TrA1730-1732 (Downton). 32 Downton’s plan hrc1C187 and Interiano’s plan strjC117 are very similar, except the El Paso configuration differs somewhat and the border between CD20 and CD23 in Bexar County differs (though the record testimony incorrectly says Bexar County is the same). TrA320 (Interiano). 485. At 1:15 p.m., Interiano emailed Denise Davis saying, “Let me know when I can come update you.” US-750. Davis responded at 2:04 p.m., saying, “Now.” 486. Downton’s RedAppl plan list includes hrc1C194 (PL-825; D-528; US-731TT), a nine-district plan titled “Solomons Floor Amendment - Final,” created at 3:16 p.m. and last modified at 3:24 p.m. on June 13. US- 658; D-508.1; PL-549. It is similar to strjC117. This plan was sent to TLC at 3:31 p.m. and became public at 4:53 p.m. as Plan C170, the Solomons West Texas Amendment. D-506. 487. Compared to Plan C149, Plan C170 returned half of Maverick County, and all of Zavala and Dimmit Counties to CD23. Plan C170 also added part of Atascosa County to CD23. TrA1691-92 (Downton). These changes are similar to changes made by Opiela in his suggested map. Downton testified that his map took Opiela’s suggestion to put half of Maverick County into CD23, but the line is not exactly the same in the City of Eagle Pass and more people are added into CD23; that his map has a different El Paso configuration because it incorporated Pickett’s suggested

32 Downton testified that Interiano created strjC116, which he imported from Opiela, and then sent to Downton, who worked on it until about 2:30 that afternoon. Around lunchtime, Downton sent a version of the plan to Interiano’s RedAppl account, which become strjC117. TrA1730-36 (Downton). 374a map for El Paso (though again the only Pickett map in evidence does not look like this and the Court does not find this testimony credible); that his map splits La Salle County but Opiela’s did not; and that both maps split Atascosa County, but his does so differently. TrA1662-63 (Downton). Downton testified that his map also kept as much of Guadalupe County whole as possible (except for Schertz, a small city that they tried to keep close to whole) based on Rep. Kuempel’s request, while Opiela’s did not. TrA1655, TrA1663. By reducing the amount of Guadalupe County in CD35, Downton was able to put more of Bexar County into CD35 to weight it more as a Bexar County district (compared to Travis County). Tr991 (Downton). 488. Downton testified that changes were made to CD20 to respect Rep. Menendez’s and Congressman Gonzalez’s wishes to put more of Central San Antonio back into CD20 and to increase its HCVAP near benchmark. However, in Plan C170 compared to Plan C149, CD35 expands to take all of downtown San Antonio and more of central San Antonio from CD20, as well as the eastern portion of the south side of San Antonio. CD20 retracts out of southeast Bexar County, and takes a central part of the south side of San Antonio from CD23 (dividing the south side among CD23, CD20, and CD35). Downton testified that because CD20 took some of the San Antonio population from CD23, that decreased the HCVAP of CD23, so they “needed to have additional Hispanic population” to maintain the CD23 HCVAP. TrA1665. 33 Downton

33 Earlier, Downton had testified that Rep. Menendez (on behalf of Congressman Gonzalez) had wanted more of downtown San Antonio (which they had moved from CD20 into CD35) to be put back into CD20, and in order to do that, they had to get additional Hispanic population into CD35 to replace what they were giving back to 20, so they took that from CD23. TrA1643. 375a testified that they could not get that population in Bexar County because it was needed for CD20 and CD35, and the part of Bexar County in CD28 was not heavily Hispanic, so they had to go south to find Hispanic population. TrA1665-66. Downton said they did not consider splitting Webb County because the Supreme Court criticized splitting Webb County in 2006, so they chose to split Maverick County. TrA1666.34 Although they could have taken all of Maverick County, it would “significantly impact the Republican performance and make it difficult for Congressman Canseco to get reelected” so they made a “political

However, between C149 and C170, CD35 expands in Bexar County to take additional population from CD20; it does not give any territory back to CD20. Thus, CD35 took population from CD20, and CD20 took population inside Loop 410 from CD23. At another time, Downton again stated that Rep. Menendez and Congressman Gonzalez testified about their concerns about the splitting up of downtown San Antonio, and mapdrawers “addressed that concern by putting it [downtown] back in 20.” TrA1753 (Downton). As noted, as the map progressed, less and less of downtown remained in CD20, and it was never put back into CD20. 34 Webb County had been in CD23 in the map used for the 1996 to 2002 elections. Tr945 (Downton); PL-305; PL-306. Downton said he did not look at the prior maps. Downton 8-31-11 depo. at 13-14. When asked at his deposition why he did not put Webb County into CD23, Downton said they were very conscious of the Supreme Court ruling criticizing the Webb County split in the last round of redistricting. He said they did not want to repeat history, so they wanted keep Webb County whole. He said if Webb County were put into CD15, then CD28 was cut off population, so only CD28 and CD23 were possibilities. He felt that CD28 was essentially based in south Texas, and Webb County is a south Texas County, so it seemed to fit well there. He stated, “I suppose you could have put it in 23 as well. But we didn’t.” Downton 8- 31-11 depo. at 13. Seliger testified that he had also considered splitting Webb County, but rejected the idea based on the Supreme Court’s decision. TrA225. 376a decision” and took only what they felt they needed to maintain compliance with § 2. TrA1666 (Downton). They also made changes in El Paso to increase the HCVAP in CD23.35 489. Downton testified that Solomons was the final approval on the Maverick split and on the entire map. TrA1668, TrA1771. However, Solomons testified that he did not know why the map was put together that way other than they were trying to address El Paso, and changes in one area cause other changes in the map. TrA1361. Solomons said there were probably different ways to draw CD23, but this was the version that was presented to him as the better version. Id. Downton testified that he was satisfied with CD20 before but Solomons was now satisfied with it (in terms of HCVAP), and they were satisfied with CD35. TrA1693. 490. Plan C170, Solomons’ West Texas Amend- ment, which affected CD11, 15, 16, 20, 21, 23, 28, 34, and 35, was made public at 4:53 p.m. on June 13. Compared to Plan C149, in Plan C170, CD20’s HCVAP increased from 60% to 62.9% (closer to the benchmark of 63.8%) and its SSVR increased from 52.3/52.8% to 55.6/56.3%. CD23’s HCVAP decreased slightly from 58.9% to 58.5% and its SSVR increased from 52.6/53% to 54.1/54.8%. CD35’s HCVAP increased from 50.2% to

35 Downton stated that in making the changes within San Antonio, they reduced the HCVAP of CD23 and needed to increase it elsewhere, “so we ended up drawing a fairly thin line across El Paso County, with the lower part being in District 23 and the upper part in District 16.” Tr920-21. The HCVAP was still lower than it had been in the benchmark, so Downton then took Dimmitt, Zavala, and part of Maverick County and put them into CD23 to keep the HCVAP at the level it had been under the benchmark. Tr918-21 (Downton). 377a 51.9% and its SSVR increased from 41.7/42.8% to 43.8/45%. The changes increased the SSVRs. TrA1338 (Solomons). Downton said that adding half of Maverick County only brought CD23’s HCVAP down by .4 and it remained above benchmark. TrA1670. In Plan C149, McCain’s performance was 51.9%; in Plan C170 it was 51.6%, a slight decrease. TrA1377-78 (Solomons); D- 692. Obama’s performance increased slightly from 47.1% in C149 to 47.5% in C170. D-692. 491. Republican performance in CD23 is slightly worse in Plan C170 and enacted Plan C185 than in Opiela’s plan strjC116 (McCain is 52.% in strjC116, 51.6% in Plan C170, and 51.6% in Plan C185). TrA1669 (Downton); TrA1551 (Hanna); D-692, D-689, D-690; D- 691. Plan C170 and Plan C185 HCVAP is 1.7% higher than strjC116’s. TrA1669 (Downton); TrA1552 (Hanna); D-677. And SSVR is higher in Plan C170 and Plan C185 than strjC116. TrA1552 (Hanna); D-677. 492. Opiela was initially unhappy with some of the changes in Plan C170 because the CD20 SSVR was lower than he wanted. TrA1666 (Downton). At 10:29 p.m. Opiela emailed Jennifer Brown (Congressman Smith’s chief of staff) and copied Smith. D-631. The email was titled “Re: okay . . . . this has to stop.” Opiela wrote, “Not that we don’t have enough to worry about, as I was making my way through all the reports for each district I finally got to the Solomons amendment. Makes me want to shoot through the roof. I didn’t go through all of this for nothing today. I got the stats in CD 20 to benchmark, 58.1 SSVR, etc. just to have them drop it back down to 55.6. I had a voice wondering in the back of my head how they were able to find enough Hispanics to jack Quico [Canseco] up from 52.8 [52.6] to 54.1. I knew they couldn’t do it alone with just 10k more in Maverick. They stole them from CD 20. This 378a was the whole point behind this exercise. I gave them the tools to fix this, and it was used for this. I’m tempted to try to get someone to offer what I gave them as an amendment to Solomons amendment, but know that will blow things up.” Lamar Smith responded to Opiela and copied Interiano at 10:40 p.m. He wrote, “Just had a long talk w harvey hildebran who called me back. He did not know about the solomon amend and when I explained it took cd20 back to benchmark and increased hispanics in cd23 he said he wld study it and barton map and might not offer barton. Said purpose of barton was to improve cd20. He also did not kno barton wld make a lot of del unhappy and hurt canseco and took three incumbents below 55 mccain so he said he wld study solomon amend tonite before making a decision. Think if burt called him and explained his amend hh wld not offer his.” D-236; D-631. 493. At 11:06 p.m., Opiela sent Interiano an email with the subject “why.” TrA1666(Downton); D-235; D- 632; NAACP-616; PL-311 Pt. 6. Opiela wrote, “Why do this to me? I get the stats to benchmark in 20 and look at the report tonight and y’all dropped them to 55! Please, please say this whole exercise was not for naught. Call me crazy but I could care less about 35 being at 45 – the thing will probably perform at 40, why not take that difference and attempt to meet it if we can? I calmed Lee, Holman, et al. down today to have this done?” D-632. 494. At 6:57 a.m. on Tuesday June 14, Interiano responded to Opiela’s “why” email with, “It’s at 56.5 [56.3] . . . . and it performs 10/10. You’re looking at total when you should be looking at non-suspense VR. Call me later this morning if you want to talk further. 379a Also, can you send me the analysis that you did on all the amendments?” PL-311 Pt. 5; D-632; NAACP-616. 495. At 6:59 a.m. on Tuesday June 14, Interiano wrote Smith and Opiela, stating “I’ll make sure to pass [the concerns raised in the “this has to stop” email] to the Chairman this morning. In re to CD20, it’s actually at 56.3 and it performs 10/10 so it’s a definite improvement from where we were and there is no reason to be concerned in that regard.” D-631. 496. Opiela responded to Interiano at 8:32 a.m., stating, “No, just I think it’s not good to not have all the stats at benchmark when we know its possible. Yes, I know CD 20’s 10 of 10 performance and agree taking it below benchmark on SSVR can be justified by the creation of the new district. However, we know we can create the new district AND keep it at benchmark, so we should. Sure it performs the same, and will it make a difference in the end with the court, probably not; it just gives an easy paragraph where they can say we retrogressed, and we have to explain why it doesn’t matter. That said, today’s amendment is enormously better than C149, so let’s just play ball. I do feel I was mislead into thinking y’all were going to keep all the districts at benchmark yesterday, and frankly, I used that argument to calm Lee and Chris Homan down when they were threatening to blow it all up. What is done is done. Let’s get this bill passed!” D-632; NAACP-616. Interiano agreed that Opiela calmed down after Interiano directed him to the right SSVR numbers. TrA376. 497. At 10:24 a.m., Opiela emailed Downton with “quico talking pts amend” with talking points for an amendment to Plan C170 offered by Peña (Plan C179) relating to CD11, CD21, and CD23. PL-311 Pt. 7 at 463 (noting that the amendment “returns Schleicher and 380a the portion of Sutton that are currently in CD11 back to CD11; and moves an equivalent amount of population from Kimble County (whole) to CD21. CDs21 and 23 are balanced in Bexar County. Schleicher and Sutton have a closer community of interest with CD11, and Kimble with CD21. It also slightly improves the SSVR in CD23.”). 498. On June 14, there was debate in the House on second reading of C.S.S.B. 4. D-23; D-603.2; Quesada- 394. The current plan was Plan C149, the plan as passed out of the HRC on June 9. Rep. Alonzo offered an exhibit into the record asserting that CD23 had been diluted. The exhibit said, “Even more egregious is the change made to District 23, which would also be controlled by high turnout Anglo precincts and counties in this plan. Although the PLAN C141 configuration would maintain its Latino population based on census data alone, a look at election data demonstrates that it would no longer be an effective Latino Opportunity District. For example, the Hispanic candidate of choice in 2008 was President Obama, who carried the district with 50.9 percent of the vote. Under this plan, McCain carried the district with 51.9 percent of the vote to Obama’s 47.2 percent––a 3.7 percent change. This outcome was achieved by exchanging selected precincts in Bexar and El Paso Counties, even going so far as to take District 23 all the way across the Franklin Mountains into the west side of El Paso to achieve a purely partisan outcome.” D-603.2 at S410. 499. Solomons offered his West Texas Amendment Plan C170, which was made public the day before, and moved passage. Solomons stated that the amendment responded to some concerns from the public and Rep. Menendez at the House and Senate hearings on the bill. He said, “Representative Menendez specifically 381a asked us to increase the SSVR of District 20 over 55 percent. We were able to increase it to 56.3 percent [non-suspense]. We also increased the HCVAP of Dis- trict 35 to 51.9 percent and the SSVR of District 23 to 54.8 percent [non-suspense]. While we were doing this, we were able to maintain the performing nature of all the other Hispanic majority districts, and it’s acceptable to the author.” D-23; D-602.3 at S367. Rep. King spoke in opposition based primarily on the fact that the amendment split Maverick County and the City of Eagle Pass and also split La Salle County. D-23 at S1-S2. 36 Rep. Villarreal stated, “this amendment is actually an improvement from what came out of committee in some very specific ways” and asked Solomons to repeat how he improved the SSVR numbers for districts 20, 35, and 23. Id. at S2. After Solomons did so, Villarreal said, “I want to thank you for making these improvements.” Id. Solomons moved passage of Plan C170, and it was adopted. Plan C170 was the final configuration for CD23. 500. Solomons testified that he did not know who drafted the Plan C170 amendment, but he was asked to introduce it, likely by staff. TrA1286. His staff wrote the statement he gave about the amendment. TrA1288 (Solomons). Regarding CD23, he said on the floor, “[W]e needed to sort of do that [increase HCVAP and SSVR] and try to maintain the performing nature of this Hispanic majority district in South Texas.”

36 Downton testified that this was the first time he became aware that he had split the City of Eagle Pass. The Court does not find this testimony credible. Even if he did not have the city overlay displayed on RedAppl, he split precincts in Eagle Pass and would have realized from the population counts in the precincts and surrounding areas that this was a more heavily populated area, such that even if he did not know the name of the city of he was splitting, he would have realized he was splitting a city. 382a D-603.2 at S367. Solomons did not know whether this was true, and acknowledged that his staff would have been aware that the OAG10 showed a decrease in Latino performance in CD23. TrA1288-89 (Solomons). 501. Between Plan C125 and Plan C170/C185, SSVR was increased in CD23, but performance for at least some Latino-preferred candidates (such as Molina in 2006) decreased. TrA663-64 (Archer); TrA1520 (Hanna). 502. On June 15, Opiela sent an email to Canseco’s staff and Smith in which he asserted that there would be a decrease in Republican votes with the proposed amendment to CD23 in Plan C170 to be offered by Rep. Peña. TrA1671 (Downton); TrA1379-80 (Solomons); D- 685; US-757. Plan C170 removed Schleicher County and part of Sutton County from CD23 (putting them back in CD11) and gave CD23 additional population from north Bexar County (taking it from CD21). Com- pared to C170, it raised the HCVAP of CD23 slightly (from 58.5% to 58.7%), while non-suspense SSVR remained the same and total SSVR dropped slightly from 54.1% to 54%. D-569.7. Opiela forwarded the information to Interiano and Bruce saying, “Pull the Peña amendment.” D-633. Bruce forwarded the email to Solomons, Downton, and Interiano, and Solomons said the amendment needed to be withdrawn and would not be accepted. TrA1349-51 (Solomons); D-685; TrA1671 (Downton). Downton said that Interiano had told him that the Republican Congressmen felt it was going to hurt performance. TrA1671. 503. Interiano remembers inquiring about and discussing whether enacted CD23 provided an opportunity for Hispanics to elect their candidate of choice, but no conclusion was reached. TrA327 (Interiano). Seliger remembered having discussions about whether 383a Latino voters in the enacted CD23 would be able to elect their preferred candidate, but stated that he could not remember what was concluded other than that he was assured by staff that the map would be a legal one. Tr239-40 (Seliger). 504. Downton testified that they did not have § 5 concerns at this point. TrA1694; Downton 8-12-11 depo. at 93-96. He felt CD23 was an opportunity district in the benchmark but that it was not performing, and that it remained the same in the enacted plan. TrA1696, TrA1800. He felt both 1/10 and 3/10 were not performing. TrA1638, TrA1696. He also felt it was an opportunity district in the enacted plan because it was over 50% HCVAP even though it only performed 1/10 in the OAG10. Tr966, TrA1800; Downton 8-12-11 depo. at 93 (stating that with HCVAP approaching 60% Latinos have the opportunity to elect). Downton felt that if he was wrong about this, any loss of CD23 would be offset by CD35 for § 5 retrogression purposes. TrA1636, TrA1696-97. Hanna testified that he never advised that CD35 could be a swap for a reduction of performance in another existing Hispanic opportunity district such as CD23. TrA1584. 505. CD23 is a swing district, so “every couple of tenths of a percent could impact an election.” TrA1669 (Downton). Murray noted that CD23 is the most closely contested district in Texas and nudging it a little bit would have more impact than in any other district. TrA1494. 506. Although there were concerns about the performance of CD23 for Latino-preferred candidates and mapdrawers had been notified of concerns from TLC counsel, Downton and the mapdrawers did not work to improve the performance of Latino-preferred candidates (such as Chavez Thompson in 2010, Yanez in 384a 2008, Molina in 2004 and 2006, and Sanchez in 2002) as the plan progressed from C125 to C185. US-761 (demonstrative exhibit based on data in Red-225 reports); US-732. From Plan C125 to Plan C185, Chavez Thompson’s performance decreased by .1% from 37.9% to 37.8%; Yanez improved from 50.6% to 50.8%; Molina (2006) decreased from 46.8% to 46.6%; Molina (2004) decreased from 47.3% to 46.8%; and Sanchez improved from 48.7% to 49.3%. In addition, the total number of voters turning out to vote in 2006, 2008, 2010 in CD23 decreased between C125 and C185. McCain/Obama performance was as follows: 51.9/47.2 (C125); 51.9/47.1 (C149); 51.6/47.5 (C185). 507. Hanna testified that he and Archer still had unresolved concerns about CD23 in the enacted plan, and the boundaries of CD23 could have been adjusted in a way that would have resolved his concerns. TrA1521, TrA1581. If he had been asked about it, he “would have expressed concerns,” though he would not have shared his knowledge about CD23’s performance dropping from 3/10 to 1/10 on the OAG 10 because of his confidentiality obligations (he did not recall being asked). TrA1572-73. 508. Canseco did not see a map of CD23 in the enacted plan until after the plan was passed, and he was not shown political performance numbers or statistics by mapdrawers. TrA576 (Canseco). His main concern was whether he could possibly get reelected, and he felt a little ambivalent about the enacted CD23. TrA576 (Canseco). He described it as “six of one, and half a dozen of the other” in terms of reflecting his goals; he was satisfied but not pleased with CD23. TrA582, TrA585-86. The enacted CD23 preserved the border and many other areas, but he lost a lot of the south side of San Antonio/Harlandale and some 385a Republican areas in Helotes. TrA574-75, TrA582-83 (Canseco). Canseco lost some of the San Antonio south side precincts that had been very strong for him in the primary (this population was swapped for El Paso territories), and he was concerned that this would make it more difficult for him if he drew a primary challenger. TrA575 (Canseco); PL-349; PL-350. Canseco also gained some West Texas areas and testified that the counties along the Pecos River that were added into CD23—Loving, Winkler, Ward, Schleicher, Crane, Upton, Frio, LaSalle—had a lot in common with the border areas because they are “heavy with oil and gas” and they are trying to have more trade and commerce go through there from the border. TrA585 (Canseco). However, the mapdrawers themselves could not offer a rationale for including these counties in CD23. Tr225-226 (Seliger); Tr963 (Downton). 509. There are 39 split VTDs in CD23: 19 are in Bexar County, 9 in El Paso County, 8 in Atascosa County, 2 in Maverick County, and 1 in La Salle County. PL-1633 (Red-381 report); US-704 (Red-110 report). Five splits (two in El Paso, one in Atascosa, and two in Bexar County) involve no population. US- 704. Many of the splits occurred toward the end of the process (between Plan C149 and Plan C185), and Downton was unable to provide explanations for the splits other than to say that some would be zeroing out and that he split counties that had not been split before (Maverick, La Salle, Atascosa), though the latter does not explain the need for or the reasons for the specific cuts. TrA1749-51 (Downton). 510. Concentrated Hispanic precincts (e.g., 1041, 1045, 1025, 1044) in the south side of San Antonio are split between CD20 and CD23. TrA1751; US- 760. Harlandale was removed from CD23 and split. 386a TrA583-84 (Canseco); Tr680 (Korbel). Downton did not know that/whether Harlandale or South San Antonio was a community of interest. TrA1752-53 (Downton). He said that even if they were, he would not necessarily need to keep them together because there are other redistricting principles besides maintaining communities of interest and they were trying to create districts under the VRA and were required to consider race to create those. While mapping, he did not check with anyone whether he was dividing communities of interest and only looked at it if someone noticed and brought it to his attention. TrA1753 (Downton). 511. Maverick County is close to 60,000 in population, and is 95% Hispanic Tr766 (Saucedo). Maverick County has never been split; it has always been in one congressional district. Tr766-767 (Saucedo); Tr447-48 (Flores). Although it had been wholly in CD23 in Plan C100, half of it is in CD28 in Plan C185. Two precincts were split in Maverick County in the City of Eagle Pass. Tr449 (Flores); PL-1633. Maverick County Judge David Saucedo testified that Plan C185 splits neighborhoods between CD23 and CD28 in Maverick County. Tr766. C185 does not respect Maverick County or City of Eagle Pass lines. Tr774 (Saucedo). Saucedo testified that Maverick County voters have been able to get behind one candidate so that they have an influence in elections, and splitting the County takes away that voting strength. Tr767-68. He testified that it was not an advantage for Maverick County to have two representatives because voters in other areas would control CD23 and CD28, and there would be no need for a representative to pay attention to Maverick County. Tr770. Maverick County would rather be a significant part of one district than an insignificant part of two districts. Tr771 (Saucedo). Maverick County produces about 12-14,000 votes in a presidential year 387a and 8-9,000 in a non-presidential year. Id. Downton testified that he first became aware that he had split the City of Eagle Pass in Maverick County when Rep. King testified about it on the House floor during the June 14 debate over the amendment (Plan C170), and by then it was too late. TrA1754. Although he was drawing at the block level in the line dividing Maverick County, he said he did not have city mark- ings visible on RedAppl. TrA1755 (Downton). He testified that if he had seen the city marking, he would have tried to keep all of the City in one district. TrA1755-56. As noted above, the Court does not find this testimony credible. Downton agreed that the Hispanic community in Eagle Pass is a community of interest, but said he did not intentionally divide it. TrA1756. Although he did not think it was a VRA violation to split Eagle Pass/Maverick County, he said he would not have done it if he had been aware of it or if he had time to fix it. TrA1755. 512. In Plan C185, significant parts of the south side of Bexar County were removed from CD23 and replaced with another part of Bexar County that had been in CD20. Tr448 (Flores). Flores noted that the South Side is a traditional voting community within CD23, bound on the west by Lackland Air Force Base and on the east by I-10, stretching all the way to Highway 1604. Joint Expert Ex. E-8 (Flores report) at 10. Flores noted that the South Side is cracked three ways in Plan C185, with the majority of it being moved into CD20 and CD35. Joint Expert Ex. E-8 (Flores report) at 11. Ciro Rodriguez testified that every Latino that gets elected in San Antonio/Bexar County needs the massive support of the Latino community to win. Tr780. He and Congressman Tejeda began forming a coalition on the south side, and the south side of San Antonio has traditionally voted as a bloc and has 388a been successful. Tr781 (Rodriguez). Rodriguez testified that the south side of San Antonio is very different from the rest of San Antonio, and has a good number of political subdivisions, including Harlandale, South San ISD, South Side ISD, Southwest ISD, Somerset ISD, and East Central. South Side has always been part of one congressional district and is a community of interest. Tr785-86. Since it was formed, candidates for commissioner’s court, city council, mayor, const- able, JP and congress have had to have support of this coalition as a bloc to win. Tr782. Rodriguez testified that it would be very difficult to win CD23 because they added ten counties, divided Maverick County, and divided Harlandale among districts 20, 23, and 35. Tr784. 513. Flores noted that, in addition to the changes in Bexar County and Maverick County, a series of counties along the northern side of the Pecos River with predominantly Anglo population that were in CD11 (Loving, Winkler, Ward, Crane, Upton, Reagan) were moved into CD23, and a significant part of El Paso was removed and another part put in from CD16, and there are precinct cuts and split communities in El Paso County.37 Tr449; Joint Expert Ex. E-8 (Flores report) at 11. Flores thought that CD23 was over manipulated, because the benchmark was overpopulated by 149,000, but more than 600,000 persons were moved around to achieve the final configuration. Tr450, TrA505-06; PL-455 at 12; PL-385; PL-399. Approxi- mately 380,000 people were moved out of CD23.

37 The counties were in CD11 in Plan C100. But in 2002, Upton and Reagan Counties were in CD23. D-654; TrA559 (Flores). In the 1992 to 2000 elections, Loving, Winkler, Ward, Crane, Upton, and Reagan Counties were in CD23. TrA559 (Flores); D-653; D- 652. Schleicher County had not previously been in CD23. 389a PL-1029; D-403. 466,974 people remained in CD23. TrA506 (Flores); PL-1029; D-403. And 231,000 people were moved in: 33,730 from CD11; 76,294 from CD16; 45,566 from CD20; 14,559 from CD21; and 61,365 from CD28. TrA506 (Flores); PL-1029; D-403. They had to remove 58,939 people from CD16, but 76,294 people were moved into CD23 from CD16 and 17,355 were then moved from CD23 to CD16. TrA507-08 (Flores); PL-399; PL-1029; D-403. 514. The percentage of Latino registered voters in areas that were brought into CD23 (considered in total) was 62.55% and the percentage of Latino registered voters in areas removed from CD23 (considered in total) was 54.12%. Tr450-51 (Flores); PL-236. This is consistent with the fact that SSVR was raised in CD23 from the benchmark. However, even though overall the areas put in had a higher SSVR%, those particular areas generally underperformed compared to those areas taken out. Tr451 (Flores); PL-236 (showing the turnout percent in 2010 for areas moved in was 30.62% and for areas moved out was 35.98%). Flores suspected that “there was a lot of very detailed looking around for VTDs that underperformed that had large numbers of Latino voters in it to be moved into the district and those areas that – VTDs that had high numbers of Hispanic voters that overperformed were removed from Congressional District 23 just to get it to have the level of Hispanic voters that it ended up having.” Tr451. Looking at Spanish Surname voter turnout, precincts in El Paso put into CD23 were all around 19% or lower. Tr452 (Flores); PL-395. Areas in Frio and Atascosa Counties put into CD23 had higher turnout rates in general. Tr452 (Flores); PL-396 (La Salle County was around 19% or lower). Areas on the south side of Bexar County that were taken out were between 25 and 30%, and areas on the northwest side 390a of Bexar County that were removed were consistently above 30%. Tr452-453 (Flores); PL-397; PL-398. 515. Although HCVAP and SSVR in CD23 increased from C100 to C185, the ability to elect the Latino preferred candidate decreased. Tr453-54 (Flores). Flores concluded that the changes were based on race given “given the way that the VTDs were manipulated, moved in and out of CD23.” Tr454. He concluded that CD23 in Plan C185 was not a Hispanic opportunity district and that a Hispanic candidate would find it very difficult to get elected in the new configuration. Tr454-55. 516. In El Paso, 32,880 registered voters, 29,043 of whom were Latino, were moved from CD16 and put into CD23, while 6,381 registered voters, 4,863 of whom were Latino (by SSVR), were moved from CD23 to CD16. The net effect was to increase the SSVR of CD23. TrA508-09 (Flores); PL-1109 at 5. However, the turnout rate for Latino voters (votes cast divided by number of SSVR) moved into CD23 was 11% while the turnout rate for voters moved out was 14.9%. TrA509 (Flores); PL-1109 at 5. It was not necessary to move higher turnout Latino voters into CD16, which was already overpopulated. TrA509 (Flores). These changes would increase the SSVR but add lower turnout Spanish-surname voters to CD23. TrA509-510 (Flores). 517. In West Texas, Loving, Winkler, Ward, Crane, Upton, Reagan, Schleicher Counties, and part of Sutton County were added to CD23 from CD11. TrA510-11 (Flores). CD11 was overpopulated by 12,194, but 33,730 people were moved from CD11 into CD23, including 19,158 registered voters, of whom 6,925 (36.1%) were Latino (by SSVR). TrA511-512 (Flores). There were almost twice any many non-Latino registered voters added, and in this part of the state, almost all non- 391a Latino voters are Anglo. TrA512-13 (Flores); PL-1109 at 3. The Latinos moved in had a turnout rate of 22.9%, whereas the non-Latino voters moved in had a much higher turnout rate of 39.6%. TrA513 (Flores); PL-1109 at 3. It creates a participation gap when a group of Latino voters with a lower turnout rate and much larger group of Anglos with a much higher turnout rate are imported. TrA513 (Flores).38 518. In South Texas, half of Maverick County was taken out of CD23, and all of Frio and parts of Atascosa and La Salle Counties added in. TrA514 (Flores). 12,322 registered voters in Maverick County were taken out, of whom 91.1% were Latino (11,230) and 1,092 were non-Latino. TrA51415 (Flores); PL- 1109 at 4. In Frio, Atascosa, and La Salle Counties, 33,487 registered voters were moved in, 21,426 (64%) of whom were Latino and 12,061 were non-Latino. TrA515 (Flores). The net effect increases the SSVR in CD23; there are roughly 10,000 more Latino registered voters coming in. TrA515 (Flores). But it also moves in 11,000 non-Latino registered voters, who are mostly Anglo, and the Latino voters moved in have turnout rate of 17.5%, while the Latino voters moved out have a turnout rate of 19.8%. TrA516 (Flores); PL- 1109 at 4. It thus brings in a group of SSVR with a

38 In both El Paso and the “West Texas” areas examined by Dr. Flores, however, the net effects were to decrease non-SSVR turnout by a larger amount than SSVR turnout. In El Paso, SSVR turnout decreased 5.2% while non-SSVR turnout decreased 10.4%. In “West Texas,” SSVR turnout decreased by 6% and non- SSVR turnout decreased by 12.6%. PL-1109. However, in the four areas Dr. Flores examined, SSVR turnout was decreased in three out of four areas (and in Bexar County it only increased by .8%), while non-SSVR turnout decreased in only two of four areas. And the overall change in the district was to decrease SSVR turnout and increase non-SSVR turnout. 392a lower turnout rate compared to the group moving out. TrA516. And the non-Latino registered voters being moved in have a high turnout rate of 41.8%, more than twice as much as the Latino turnout rate, while the non-Latinos being removed had a rate of 22.8%. TrA516-17 (Flores); PL-1109 at 4. This widens the participation gap. TrA517 (Flores); PL-1109 at 4 (SSVR turnout is down by 3.7% while non-SSVR turnout is up by 11.3%). 519. In Bexar County, 173,405 registered voters were moved out of CD23, of whom 86,433 were SSVR, and 24,782 registered voters were moved into CD23, of whom 10,469 were SSVR. TrA51718 (Flores); PL-1109 at 6. This reduced the population of Latino/SSVR by about 76,000. TrA518 (Flores). The turnout rate of Latino voters moved out of CD23 was 23.9%, and the turnout rate of Latinos moved in was 23.5%, which is similar, but it still removed 76,000 Latino voters with consistent voting patterns. Id. In Bexar County, SSVR turnout was increased only .8% compared to an increase of 4% in non-SSVR turnout. PL-1109 at 6. 520. None of the changes in those areas improved Latino voting strength in CD23. TrA518-19 (Flores). Higher SSVR turnout areas in Bexar and Maverick Counties were removed and lower turnout areas in El Paso and from CD11 were included. TrA519 (Flores). Looking at the district as a whole, the effect was to reduce the SSVR turnout rate in November 2010 from 24.9% to 22.9%, a decrease of 2% (even though the total SSVR of the district increased by 2.1%), while at the same time increasing the non-SSVR turnout by 2.1% from 39.4% to 41.6%. TrA519-20 (Flores); PL- 1109 at 2. This is a significant change for CD23, in light of how close some elections have been historically. TrA520 (Flores). In CD23 as a whole, the SSVR 393a turnout rate in C185 is 22.9%, and non-Latino turnout rate is 41.6%, almost double. TrA523-24 (Flores). In the benchmark, SSVR turnout was 24.9% and non- Latino turnout was 39.4%. PL-1109 at 2. Dr. Flores opined that the numbers match with the nudge factor and that there was an intent to depress the Latino turnout rate while giving the impression that CD23 was a Hispanic opportunity district. TrA525. On cross-examination, Flores agreed that Opiela proposed decreasing a district’s SSVR, while mapdrawers increased SSVR, and he agreed that the spread between HCVAP and SSVR actually decreased. TrA539-40; D- 67239; D-677. 521. All of Dr. Flores’s turnout data was from the 2010 election. TrA548 (Flores). He did not use 2008 turnout data even though turnout was higher. TrA548-49 (Flores). He also did not look to see which precincts being moved in and out were “better for Republicans.” TrA551-52 (Flores). Flores did not do a reconstituted election analysis of CD23. TrA560. Dr. Flores acknowledged that Latino turnout actually increased in Bexar County portions of CD23 (by .8%). TrA555 (Flores); PL-1109 at 6. But he noted that it is only from 24.5% to 25.3%, and the number of SSVR voters is reduced 75,964, such that SSTO is up by .8, but the number of SSVR voters is down by 75,964. TrA562.

39 D-672 contains an error. It shows that the gap between HCVAP and SSVR decreased from 6.4 in Plan C100 to 3.7 in Plan C185 (a decrease of 2.7), but it erroneously uses the non-suspense SSVR for Plan C185 instead of total SSVR while using total SSVR for Plan C100. When using only total SSVR, the correct difference is a decrease of 2 (from 6.4 in Plan C100 to 4.4 in Plan C185). When using only non-suspense SSVR, the gap decreased from 5.8% in Plan C100 to 3.7% in Plan C185, a decrease of 2.1. D-677. 394a 522. In the OAG’s “Estimated Turnout by Race/Ethnicity as a Percent of VAP in VTDs” in CD23, estimated turnout percent for Anglo voters is increased in all ten Democratic Primary and General Elections looked at, the total increase in Anglo voter turnout percent in the general elections is 32.3%, and the average Anglo turnout percent increase in the general elections is 6.46%. In contrast, the estimated turnout percent for Hispanic voters is increased in only eight of the ten Democratic Primary and General Elections looked at, the total increase in Hispanic voter turnout percent in the general elections is .8%, and the average Hispanic turnout percent increase in all the general elections is .16%. US-600; US-601; Quesada-271. 40 This estimated turnout data was available to the mapdrawers. 523. In Plan C100, the Latino-preferred candidate won in 3/10 elections in the OAG10, while in Plan C185, it drops to 1/10. TrA521 (Flores); PL-1130; TrA1637-38 (Downton). 524. Dr. Alford noted that benchmark CD23 elected a Democrat in 22/48 (46%) of statewide elections while the new CD23 wins 14/48 (29%). In the 2011 trial, Dr. Alford agreed that Latino performance decreased and that it was probably less likely to perform than the benchmark, and stated that he would not count CD23 as a performing district. Tr1839, Tr1878 (Alford). He also would not recommend changing CD23 the way it was changed. Tr1838, Tr1879 (Alford).

40 DOJ asserts that Hispanic turnout decreased, but it may be considering the total number of votes cast in CD23 in Plan C100 compared to C185. Because of the population differences in the district in Plan C100 and C185 (Plan C100 was overpopulated), the Court does not find this to be a reliable comparison. 395a 525. In addition, CD23 in Plan C185 is not safer for Canseco in a racially contested primary. Tr455 (Flores). Canseco won the primary runoff in 2010 by less than 800 votes. Tr455-56 (Flores). Canseco got his highest votes in the Republican primary runoff in Bexar County and a little part of Zavala County. Tr456 (Flores); PL-350. Similarly, Carillo (incumbent) got his highest Republican primary returns in Bexar County and a little in Medina. Tr456; PL-381. The changes made to Bexar County would make it much more difficult for a Hispanic Republican in a racially contested primary. Tr457 (Flores). The impact of changes to CD23 means that in a racially contested Republican primary, a Hispanic candidate could not win or would have an extremely difficult time. Id. Support for Latino Republicans in the primary decreased from Plan C100 to Plan C185 in CD23. PL-200. 526. Dr. Arrington concluded that there was evidence of the nudge factor being used in CD 23. TrA397- 98. Arrington looked at all precincts/blocks taken out and precincts/blocks put in between the benchmark and new CD23. TrA399. He looked at the turnout level for the precincts moved in and out to see whether the district was shaped to have lower Hispanic turnout while at the same time maintaining demographics. TrA401-02. He determined that in the heavily His- panic VTDs (those with more than 65% Hispanic population) that were moved out, the 2010 turnout level was 16.1%, and in the heavily Hispanic VTDs that were moved in, the turnout level was only 12.7%. TrA401-02. In other words, the VTDs moved in had significantly lower turnout than the ones moved out. TrA403 (Arrington). He found that the exogenous election index indicates that the new CD23 offers no opportunity. TrA405. Arrington agreed that Opiela’s nudge factor email suggested keeping the SSVR low, 396a but Plan C185 increases the SSVR and narrows the gap between HCVAP and SSVR. TrA442-43. Arring- ton concluded that the purpose of the reconstruction of CD23 was to give an incumbent Republican who was not the Latino candidate of choice a better chance of being re-elected, that precincts moved into CD23 had higher levels of Republican support than precincts moved out, and that the goal was met. TrA450-51. 527. Lichtman testified that one cannot judge CD23 just by its demographic statistics and that CD23 in Plan C185 does not provide a reasonable opportunity to elect minority candidates of choice. Tr1235-36. The Latino candidate of choice lost all five elections in his study, some by wide margins (Perry beat White by 10 points; Dewhurst beat Chavez-Thompson by 19 points). Tr1236 (Lichtman); Joint Expert Ex. E-3 at 17. Elec- tions were closer in 2008 but the pattern was the same (McCain defeated Obama by about 4 points; Cornyn defeated Noriega by about 2.5 points, and Jefferson defeated Jordan by about 2.5 points). Tr1236. Lichtman opined that low turnout Latinos were put in and very high turnout Anglos were put in, who strongly vote against the Latino candidate of choice, so it is a combination of effects (turnout and degree of Anglo bloc voting) that explain why one can never judge a district by its cover alone. Tr1258. 528. Dr. Alford testified that CD23 in Plan C185 provides Hispanic opportunity because both the HCVAP and SSVR are increased and are significantly over 50%, so Hispanic registered voters have an opportunity to control the vote. TrA1849, TrA1859. In his report, Alford says performance is not relevant to a § 2 analysis. D-424 at 6. He testified that looking at the whole geography of CD23, EI analysis shows that the Hispanic vote breaks majority Democrat and typically 397a the Democrat candidate is the candidate of choice. TrA1851. However, CD23 “tilts Republican” because it has elected a Republican at times and based on exogenous election results. TrA1851 (Alford). He noted that CD23 in Plan C185 is undoubtedly a slightly more Republican tilt than it was in C100, and the version in the interim Plan C235 is somewhere in between the two. TrA1850. Alford testified that the 2012 elections show that it is capable of electing a Democrat even in a year when it tilts Republican. TrA1851. He noted that, in Plan C235 in 2012, the top of the ticket exogenous elections are majority Republican (voting for Romney and Cruz), but it nevertheless tilts Demo- cratic in endogenous elections, so Democrat Gallego was elected. TrA1851-52. Alford found this also in Plan C100 in 2006—despite the exogenous election pattern tilting Republican, the Hispanic preferred candidate can still be elected in CD23. TrA1852. To determine whether this would have also been true under Plan C185, he looked at the 2012 election data under Plan C235 and did “a reconstituted mosaic” by summing up the Republican and Democrat votes from VTDs that would have been in CD23 in Plan C185 but that were in other districts (and thus had other congressional candidates) in Plan C235. TrA1852 (Alford); D-424 (Alford supp. report); D-428 (Table 4).41 The district was reconstructed at the block level, and it showed that the vote for the candidates of choice (O’Rourke, Castro, Gallego, and Cuellar) adds up to a higher number than for the Republican opponents. TrA1854.42 Based on this, Alford concluded that it is

41 Alford has not seen any other experts using this mosaic analysis and has not seen it in any reported cases or scholarly publications. TrA1917-18 (Alford). 42 Alford acknowledged that the candidate-specific and race- specific issues would then be incorporated into the analysis. 398a possible that Gallego would have been elected under the C185 configuration of CD23. TrA1855 (Alford). Alford did the same thing with the 2010 election using the C100 district configuration (as shown in Table 1), and Canseco would have won, and thus he concluded that if elections had been held in the C185 configuration of CD23 in 2010 and 2012, the outcomes could have been the same as they were under C100 in 2010 and C235 in 2012. TrA1856 (Alford); D-425.43 In other words, he concluded that in Plan C100 and Plan C235 there is the potential for the endogenous election to be more favorable to candidates of choice than the exogenous pattern would suggest on the surface, and his analysis suggests that it is at least possible that the same would be true for the C185 configuration. TrA1857. Dr. Handley was critical of Alford’s “mosaic” analysis of CD23 because it departs from the idea that one should use statewide elections that are the same across all components of the elections you are looking at; he borrowed from different congressional elections with different candidates and different dynamics and compiled them. This methodology did not indicate to her how a minority-preferred candidate would do in that district. TrA595 (Handley). 529. In Table 6, Dr. Alford performed a reconstituted exogenous election analysis of six elections for

TrA1882. But he felt that comparing congressional elections was a more appropriate comparison than comparing a congressional election with a different election such as a presidential election. Id. His preference would be to perform an exogenous election analysis. TrA1883. 43 In Table 5, Alford looked at how the district would have performed in 2012 if they had left the same geography as in Plan C100, and it suggested that Gallego would have been elected by a slightly more secure margin than he was actually elected. TrA1856 (Alford); D-429. 399a CD23. TrA1857 (Alford); D-430. In an analysis of six statewide exogenous elections, in C100 the Democrat wins 2/6 elections; in C185, the Democrat wins 0/6 elections; and in C235, the Democrat wins 2/6 elections. This replicates a table produced by Ansolabahere, which shows that in C185 the Republican tilt in CD23 is stronger than in C235, but adds in Plan C100 to show that it has the same pattern as C235. TrA1857- 58 (Alford). Alford testified that the results are consistent with an effort to increase Republican performance in CD23 under Plan C185. TrA1858-59. Alford opined that the character of CD23 as a majority- HCVAP and majority-SSVR but politically competitive district did not change across Plans C100, C185, and C235, but that it is simply a matter of degree. D-424 at 13. He agreed, however, that the performance of CD23 for Latino-preferred candidates is measurably worse in C185 than C100, that this has been clear from the beginning, and that it is self-evident, given the stated purpose of redrawing CD23 to improve its Republican performance. TrA1887. He also agreed that CD23 is cut very close and “it is possible to make small changes to that district that would render it less likely in a given election to elect a Latino-preferred candidate.” TrA1888. He agreed it is possible to reduce the likelihood that a Latino-preferred candidate might win by selecting precincts on the basis of turnout. TrA1888-89. 530. Dr. Handley disagrees with Alford that CD23 is substantively very similar in C100 and C185. TrA589. She notes that the minority population percentages (HVAP, HCVAP, SSVR) in C100 and C185 are similar, but the composition of a district is only a starting point for making a determination of whether a district has an ability to elect, and actual election performance 400a must be considered. US-687A at 2. Using her functional analysis, she concluded that the C100 version of CD23 provided opportunity to elect and C185 version does not. TrA590 (Handley); US-687A (Handley 4-28- 14 report).44 In C100, the minority candidate prevailed in 2/3 endogenous elections, and in 3/6 of exogenous elections. TrA595, TrA617. 45 In C185, there are no endogenous elections, and the exogenous index shows

44 Handley uses a functional, election-based approach to determine whether a district provides an opportunity to elect. TrA590. She does a racial bloc voting analysis to determine voting patterns by race and looks at recompiled election results with crossover, cohesion, and participation. TrA590. There is no particular demographic threshold, including CVAP and SSVR, for determining opportunity, but she does look at demographics as a starting point. TrA591, TrA600. She created two indices— endogenous index and exogenous index. Endogenous is actual elections in the district. TrA592. Actual endogenous elections are most relevant. TrA600-01, TrA618. Handley believes that 50% is on the line of ability to elect, and it would be preferable to have more information; she looks for ability to elect more than half the time. TrA600-01. Texas provided RPVA for a number of statewide elections and she also did her own RPVA, and a 2012 RPVA. TrA592. For exogenous elections she used six statewide top of the ticket elections that included a minority-preferred candidate. TrA593, TrA611-13. She looked at one election per year because each election cycle is different, and recent elections are more probative than older ones. TrA593, TrA610. To make inferences about a district it must be assumed that the level of minority turnout will be roughly the same in the future as it was in the past. TrA608. Outcome also depends on cohesion and the level of crossover voting. TrA608-09. Texas is very polarized and she would not anticipate it becoming less so in the near future. TrA609. 45 Handley’s 2014 exogenous election index includes six general elections, the 2002 election for Governor, the 2004 election for Court of Criminal Appeals, Place 6, the 2006 election for Lt. Governor, the 2008 election for U.S. Senate, the 2010 election for Lt. Governor, and the 2012 election for President. 401a 0/6 wins. TrA596. Handley finds no evidence that CD23 in Plan C185 would provide ability to elect. Handley uses “opportunity to elect,” “ability to elect,” and “effective” interchangeably. TrA599, TrA616. 531. Alford analyzed the nudge factor claims, but only in the Bexar County precincts moved in and out of CD23. TrA1860, TrA1871. Alford took Engstrom’s 2010 data and summed it up based on geographical location (which Flores did not do). TrA1860-61; D-431. Thus, he found the 2010 SSTO (measured by SSTO divided by total SSVR) of all “South Central” Bexar County precincts moved out was 23.3%; the SSTO of “West” Bexar County precincts moved out was 29.9%; the SSTO of all “Far North” precincts moved out was 39.3%; and the SSTO of the “Far South East” precinct moved out was 17.9%; while the SSTO of the “Lackland” precincts moved in was 21% and the SSTO of the “Far North” precincts moved in was 44.%. D-431. In total (and not counting split precincts), the SSTO as a percentage of SSVR for the Bexar County precincts removed from CD23 was 25.8%, compared to 25.4% for precincts added. TrA1861; D-424 at 15. His conclusion was that, “Despite the fact that on the map [Flores] presented it looks as though there is a substantial difference in the turnout rates, in fact, the turnout rates are virtually identical between the VTDs that were taken out and the VTDs that were moved in.” TrA1861. Alford concluded that, with regard to Bexar County, the mapdrawers did not remove high His- panic turnout VTDs and move in low Hispanic turnout VTDs in Plan C185. TrA1862.46 Alford acknowledged

46 This comports with Dr. Flores’s analysis, which showed that in the Bexar County portions of CD23, the SSTO% increased from 24.5% in Plan C100 to 25.3% in Plan C185, an increase of .8%. TrA555 (Flores); PL-1109 at 6. However, Dr. Flores noted that 402a that in Bexar County, many more people were moved out of CD23 than were moved in. TrA1875. 532. Dr. Alford testified that to make a district safer for a Hispanic Republican, he would make the district more Republican to win the general election and more Hispanic (more SSVR) to win the primary. TrA1913- 14. These are in tension and each goal places limits on the other. TrA1914-15 (Alford). Alford opined that the increase in SSVR is inconsistent with the notion that this was just an attempt to make the district more Republican. TrA1916. Alford also agreed that it would be useful to know the likelihood that Anglo Republi- cans in any particular geography would support a Hispanic Republican candidate in the primary when drawing such a district. TrA1925-26. 533. With regard to Alford’s nudge factor analysis, Handley had some criticisms of his methodology and his conclusion that the mapdrawers failed to decrease SSTO as a percentage of SSVR; she notes that there is no evidence that the data Alford used to calculate SSTO as a percent of SSVR were available to the mapdrawers. TrA622. Handley asserts that they could not have looked at SSTO to intentionally manipulate SSTO as a percentage of SSVR without having the SSTO data. TrA623-24. Handley does not find SSTO particularly relevant because it looks at how many

although there was an increase in SSTO, the total number of Spanish Surname registered voters in Bexar County was decreased by 76,000. His exhibit also shows a decrease of non- Spanish Surname registered voters of approximately 73,000, but the non-Spanish Surname TO rate increased by 4% from Plan C100 to Plan C185. PL-1109 at 6. Flores’s conclusion was that the overall effect of areas moved in and out of CD23 was to decrease SSVR turnout, increase non-SSVR turnout, and widen the turnout gap. 403a registered voters turned out and not how many failed to register in the first place, and both are components of Hispanic participation. TrA624-25. But she agreed that, if she were trying to determine whether a district were gaining or losing Hispanic voting population, she would look at SSVR. TrA631. 534. With regard to the “nudge factor” claims, Handley divided the VTDs into three groups—those that remained in CD23, those moved out of CD23, and those moved into CD23. She compared the VAP and used that in conjunction with registration and turnout rates (total registration as a percent of total VAP, turnout as a percent of registration, and turnout as a percent of VAP for all VTDs and for “heavily Hispanic VTDs”) to compare the relative participation rates of VTDs moved in and out. TrA596, TrA626 (Handley); US-678A. This information was available in various RedAppl reports. TrA597 (Handley). Handley found that the VTDs that were moved out of CD23 had higher participation rates than the VTDs that were moved into CD23. TrA597; US-687A (Table 5). She was interested in determining whether they were sub- stituting Hispanics who participated with Hispanics who did not. TrA597 (Handley). In both 2008 and 2010, participation rate of Hispanics (measured as turnout as a percent of VAP) in heavily Hispanic VTDs47 was higher in the VTDS moved out (29.5% in 2008 and 16.1% in 2010) than in the VTDs moved in (28% in 2008 and 12.7% in 2010). TrA598. She agreed that this would have contributed to the loss of the ability to elect minority preferred candidates. Id. However, she also found this turnout difference when she looked at all VTDs moved in and out, not just the heavily Hispanic

47 Handley did not look at SSTO, and she assumed that heavily Hispanic VTDs reflected Hispanic behavior. TrA627-29 (Handley). 404a VTDs. TrA633. In fact, the differences were greater for all VTDs (which includes heavily Hispanic VTDs) than for just heavily Hispanic VTDs. TrA635-37. 535. Dr. Engstrom concluded that CD23 was not a Latino opportunity district in Plan C185 (and Engstrom considers Latinos to have a reasonable opportunity in a district when their preferred candidates win a majority of the votes cast more often than not in reag- gregated elections). Tr515-16 (Engstrom); Engstrom Corr. Rebuttal Report (Joint Expert Ex. E-8) (docket no. 307-1) at 28. Ansolabahere agrees that CD23 is likely not a performing district because the vote share won by minority preferred candidates is low; it is questionable. TrA941 (Ansolabahere). 536. Canseco did not have a Republican primary opponent in 2012 under Plan C235, but he was defeated in the general election by Democrat Pete Gallego. Gallego won with 50.3% of the vote, defeating Canseco by 4.75%. TrA522 (Flores); PL-1034. Canseco had two Republican primary opponents in 2014, and Canseco did not win (Will Hurd, a non-Latino won). TrA577 (Canseco). Hurd then narrowly defeated Gallego in the 2014 general election. 537. Between Plan C100 and Plan C185, the HCVAP of CD23 was increased from 58.4% to 58.5%, total SSVR was increased from 52% to 54.1%, and non- suspense SSVR was increased from 52.6% to 54.8%, such that the difference between HCVAP and total SSVR decreased from 6.4 to 4.4 (a decrease of 2) and the difference between HCVAP and non-suspense SSVR decreased from 5.8 to 3.7 (a decrease of 2.1). Performance for Chavez-Thompson in 2010 decreased from 41.7% to 37.8%; performance for Yanez in 2008 decreased from 53.7% to 50.8%; and performance for Molina in 2006 decreased from 49.6% to 46.6%. US- 405a 761. Performance for Obama in 2008 decreased from 50.9% to 47.5%. D-692. Republican performance in statewide elections increased. TrA1858-59 (Alford); D- 430. 538. Benchmark CD23 was a Latino opportunity district. 539. CD23 in Plan C185 remained above 50% HCVAP (58.5%) but its performance for Latino-preferred candidates was significantly decreased. It was no longer an opportunity district. 540. Mapdrawers redrew the district to protect a candidate that they knew was not the Latino candidate of choice. 541. Although HCVAP and SSVR in CD23 increased from C100 to C185, the ability to elect the Latino- preferred candidate decreased, and the mapdrawers knew this was true and intended this outcome. Mapdrawers felt that they could intentionally depress Latino ability to elect as long as they kept HCVAP and SSVR numbers above benchmark. They knowingly decreased performance for Latino-preferred candidates by intentionally manipulating voter turnout and Hispanic voter cohesion. Mapdrawers claimed to be giving Canseco the best chance to be re-elected as a Hispanic Republican, but did not take steps to ensure that his primary performance was maintained or increased, and in fact made it harder for him and other Hispanic Republicans to win a racially contested Republican primary. They looked for population to include in CD23 that would raise the SSVR and HCVAP of the district but decrease its performance for Latinos, and falsely claimed on the House floor that the changes were to protect Latino performance in the district. 406a 542. Between Plan C100 and C185, SSVR turnout in the 2010 general election in CD23 decreased by 2%, and non-SSVR turnout increased by 2.1%; therefore, the turnout gap increased by 4.1%. African-American Congresspersons 543. The benchmark Congressional plan C100 included three African-American districts, each represented by African-American congresspersons. These three districts remained in Plan C185. However, the African-American congresspersons assert that the mapdrawers’ treatment of the African-American districts CD9 (represented by Congressman Al Green), CD18 (represented by Congresswoman Sheila Jackson Lee), and CD30 (represented by Congresswoman Eddie Bernice Johnson), including the unnecessary creation of tension between Hispanic and African-American residents in those districts and the questionable future viability of the districts as African-American districts, the removal of “economic engines”48 and homes and offices, and the lack of input of the African-American

48 Johnson stated that an economic engine is “a positive type of development that adds to the quality of life” and could include large employers, corporate headquarters, colleges/universities, sports stadiums, airports, hospitals, parks, historical landmarks, and neighborhoods. TrA710-11. She agreed that opinions may differ on what qualifies as an economic engine. TrA711. She testified that economic engines add to the quality of life of constituents, that a member can work with the business community to help constituents, and that removing economic engines impacts voters. Tr1521, TrA711, TrA729 (Johnson). Dr. Murray testified that removal of economic assets diminishes the member’s opportunity to leverage such assets for the benefit of a district that contains a high percentage of persons below the poverty line and families without health insurance. Joint Expert Ex. E-4 (Murray Report) at 28. 407a congresspersons into their districts, is evidence of intentional discrimination. TrA1439, TrA1412 (Murrray). 544. In Plan C100, CD9 was 36.7% Black and 42.4% Hispanic in terms of total population, and 36.3% Black and 38.9% Hispanic in terms of VAP. It was 48.2% BCVAP and 19.1% HCVAP. 545. In Plan C185, CD9 was 38.3% Black and 38.8% Hispanic in terms of total population, and 37.6% Black and 35.8% Hispanic in terms of VAP. It was 47.5% BCVAP and 18.3% HCVAP. It retained 84.1% of its benchmark population. 546. In Plan C100, CD18 was 37.6% Black and 43.5% Hispanic in terms of total population, and 37.9% Black and 39% Hispanic in terms of VAP. It was 46.4% BCVAP and 22.3% HCVAP. 547. In Plan C185, CD18 was 40.7% Black and 36.4% Hispanic in terms of total population, and 40.5% Black and 31.9% Hispanic in terms of VAP. It was 48.6% BCVAP and 17.4% HCVAP. It retained 67.4% of its benchmark population. 19.9% of its population came from benchmark CD9. 548. In Plan C100, CD30 was 42.4% Black and 39.7% Hispanic in terms of total population, and 42.5% Black and 34.7% Hispanic in terms of VAP. It was 49.8% BCVAP and 19.8% HCVAP. 549. In Plan C185, CD30 was 45.6% Black and 40.3% Hispanic in terms of total population, and 46.5% Black and 35.6% Hispanic in terms of VAP. It was 53.1% BCVAP and 20.6% HCVAP. It retained 80.7% of its benchmark population. 550. The African-American districts in Plan C100 were close to ideal population and did not require much modification. TrA1387 (Murray); TrA686-88 408a (Johnson); Tr1336 (Green). CD9 was approximately 35,000 overpopulated; CD18 was approximately 22,500 overpopulated; and CD30 was approximately 8,000 overpopulated. Dr. Murray opined that the African- American districts were substantially altered unnecessarily, and the respective representatives felt similarly. TrA1442. 551. In September 2010, Rep. Todd Hunter, Rep. Aaron Peña, Texas House Speaker Joe Straus’s legislative director Lisa Kaufman, Tom Phillips of Baker Botts, and a representative from the Texas OAG’s office went to Washington, D.C. and met with members of the Texas Congressional delegation, including Congresspersons Eddie Bernice Johnson, Al Green, and Sheila Jackson Lee. TrA683 (Johnson); TrA1079 (Hunter). The purpose was to make the congressional representatives aware that redistricting would be in the next year, let them know about the field hearings, and give them an opportunity to provide any input to the Redistricting and Judiciary Committees. TrA1079 (Hunter). Hunter told the congressional members and staffers that they could communicate with him. TrA1080 (Hunter). Congressman Smith was designated as the coordinator for the congressional delegation’s map drawing. TrA683 (Johnson). Opiela was designated as the attorney for the delegation and the person drawing maps for the delegation. TrJ1478-79, TrJ1615 (Interiano); TrA683 (Johnson). Green, Jackson Lee, and Johnson believed there would be additional delegation meetings, but there were no further delegation meetings that included them. Tr1351-52 (Green); Tr1516-20 (Jackson Lee); TrA728 (Johnson). 552. On January 11, Clare Dyer of the TLC sent a letter to each of the members of the Texas Senate, Texas House, 112th Texas Congressional Delegation, 409a and SBOE notifying them that they planned to include the census block location of their residences (rather than their actual residence address) in RedAppl and asking them to notify TLC if the included address and census block information were incorrect by January 26. D-503; D-224 (with information for Eddie Bernice Johnson). TLC staff faxed the letter because it was most expedient. TrA776 (Dyer). The TLC does not have a fax confirmation sheet. TrA767 (Dyer). The letter to Congresswoman Eddie Bernie Johnson contained her correct address,49 but the shaded census block depict- ing the location of her residence in RedAppl was incorrect. D-504. It was based on 2009 geography because TLC did not have the 2010 census geography yet. TrA779 (Dyer) (noting that it incorrectly states “2010 census block” in the corner). 553. Ten responses were received from the congressional delegation. TrA777 (Dyer). However, Congress- woman Johnson did not receive the fax and did not respond, and the erroneous census block information was not identified and corrected. TrA725 (Johnson);

49 Addresses were obtained from the candidates’ filings for office, and then verified with a second source. TrA777 (Dyer). D- 503 at 102 shows that TLC verified Congresswoman Johnson’s address with the Dallas Central Appraisal District and a white pages online listing. TrA778 (Dyer). There were a number of documents that identified Congresswoman Johnson’s residence and office that would have been available to the State. TrA697- 99. It is undisputed that TLC knew Johnson’s correct home address. TLC did not provide the actual physical addresses to members of the Legislature. TrA790 (Dyer). Interiano also obtained a list of congresspersons’s residential addresses, including Johnson’s, from the Texas Secretary of State’s office on January 6, 2011, but it is unknown whether he shared this with Downton. D-258. 410a TrA767 (Dyer). Dyer did not learn that Congress- woman Johnson’s residence was located in the wrong census block until her deposition in this case. TrA780 (Dyer). An individual working in RedAppl would not know it was wrong unless they had a reason to double check the address. TrA782 (Dyer). There is no evidence that any mapdrawer knew that Johnson’s address was shown in the incorrect census block in RedAppl. 554. Using the find-address feature, a mapdrawer can enter an address in RedAppl and then see which district it is in a particular map. TrA774 (Dyer). If someone had put in Johnson’s home address, it would have shown it (and what district it was in) on the map. Id. 555. Mapdrawers could also use RedAppl to identify points of interest such as universities or corporate headquarters by typing in a known address or with a Google Earth overlay feature. TrA782-83 (Dyer). The Google Earth feature allows a RedAppl user to take a plan and overlay it on Google Earth and then zoom in to see the district lines and features. TrA783 (Dyer). A user cannot draw a map with the Google Earth overlay feature showing. Id. The Google Earth feature is cov- ered in the training as one of the extra features discussed in the last hour, so not everyone learns about the feature. Id. A mapdrawer would not be able to see the button for the Google Earth feature unless they knew to look for it. TrA784 (Dyer). 556. TLC has never looked for any member’s district office for redistricting or entered district offices in the RedAppl system. TrA772-73 (Dyer). District offices were not shown in RedAppl. Tr1020-21, TrA1675-76 (Downton). 411a 557. Congresswoman Johnson was asked by the congressional delegation to put together a map for her area, and she met with Texas legislators Sen. West, Rep. Veasey, Rep. Giddings, Rep. Caraway, and Rep. Johnson about it in February 2011 because there were no other Democrat congressional representatives in her area. She also later met with Rep. Y. Davis and former Rep. Domingo Garcia. TrA684-85 (Johnson). Johnson testified that she presented her proposed map for her district to this group, which included a number of African-American legislators. TrA685. 558. On March 20, 2011, Congresswoman Johnson’s office sent a proposed map of CD30 with the address and location of her home, her district office, and several specific desired precinct inclusions (including the portion of a precinct including Love Field) and exclu- sions to Congressman Lamar Smith and Opiela. TrA686, TrA728 (Johnson); NAACP/AA-75. Her proposed district does not differ drastically from the benchmark CD30. Johnson expected her district map and preferences to be passed on to state legislators, but there is no evidence indicating that it was. The proposed CD30 does not match the CD30 in the map sent to Downton by Opiela on April 6 (hrc1C104). D- 510.1. Downton testified that he did review an overall map for DFW submitted by Johnson, which included her own district, but it is not clear what map he is referring to and thus whether it included the information concerning her home, office, and desired features. Tr1019 (Downton). 559. Early in the redistricting process, Sheila Jackson Lee went to Austin and met with Seliger and Solomons; she tried to make Solomons aware of the history of her district and its communities of interest. Tr1517, 1520 (Jackson Lee); Tr1627, TrA1320 412a (Solomons). Solomons was aware that she was generally satisfied with her benchmark district. Tr1627-28 (Solomons). 560. Downton testified that when he was drawing, he did not know about specific “points of interest” or “economic engines” and he did not know RedAppl allows users to see where a university, corporate headquarters, or the like are located. TrA1675. He stated that he did not intentionally include or exclude such features, and if a certain congressperson wanted a university, business headquarters, or other “economic engine” in their district, they or their staff would contact him to let him know. Id. He did not take features from any member intentionally, “other than in the context of if someone requested and they got it, well then someone else didn’t get it.” TrA1676 (Downton). Downton did not intentionally exclude members’ district offices because he did not know where they where. TrA1676 (Downton). Because Downton did not pay attention to district offices, he unknowingly removed offices from Anglo and African-American districts. Downton would only find out if a proposed map excluded a member’s office if the congressperson or their staff called him or reported it to a representative who called him, and that was the only time they looked at district offices. Tr1020-21, TrA1676 (Downton). 561. After the first public plan (C125) was released on May 31, Congresswoman Sheila Jackson Lee (CD18) issued a statement on June 2 opposing the plan. Tr1518 (Jackson Lee); D-664; NAACP-608. Solomons did not recall having seen the statement during the special session (even though it was discussed on the floor) and testified that Jackson Lee did not contact him between June 2 and June 15. 413a TrA1373, TrA1377 (Solomons). Jackson Lee had never seen the map or her proposed district before the first plan was released, and she had no input in the final result. Tr1517-20 (Jackson Lee). She was disappointed that the plan did not reflect the major minority growth, and stated that in her district, CD18, “historic neighborhoods are divided and major communities of interest have been carved from the district.” D-664. At the June 2 HRC hearing on Plan C125, Rep. Alvarado (Hispanic, Democrat) submitted Jackson Lee’s statement in opposition to Plan C125 into the record, and she also noted that Congressmen Al Green opposed the map. D-601 at 2. Jackson Lee felt that the mapdrawers had no interest in listening to her input. Tr1511. 562. Congresswoman Johnson testified that her home was no longer in her district in Plan C125. TrA686 (Johnson). However, using the find-address feature, it appears that her home was in CD30 in Plan C125, but changes were made along the border of CD30 and CD32 in Plan C130, and Johnson’s home was then placed into CD32. Because her home was shown in an erroneous census block, however, Con- gresswoman Johnson’s residence appeared in RedAppl to still be in CD30 even though it was not. TrA725 (Johnson). Johnson’s home also appears to be in CD30 in the enacted Plan C185. TrA725-26 (Johnson); TrA781-82 (Dyer); D-496. Johnson testified that she did not learn that her home and office were out of her district until the map was adopted and had become law. Tr1306, TrA686. She contacted Congressman Lamar Smith, Opiela, and her attorney about it. TrA686 (Johnson). There is no evidence that mapdrawers became aware during the redistricting process that they had not included Johnson’s home in her district or that anyone asked them to remedy the problem 414a before the map was adopted. 50 Tr1019 (Downton); TrA1373 (Solomons). 563. Mapdrawers did not add any new minority districts in the Houston or DFW areas, but new Anglo- majority districts CD33 and CD36 were added around the metro areas, causing existing districts in the area to shift somewhat. TrA1675 (Downton). However, CD33 does not enter Dallas County, CD30 remains in the same general location, and there is no indication that the addition of CD33 caused any of the changes to

50 Plaintiffs submit proposed fact findings that Johnson “specifically and directly communicated with the drafters of the map that her home and office were left out of the proposed district and she wanted that corrected.” Docket no. 1281 (proposed fact finding 54). However, Johnson testified that after she turned her DFW area map over to Smith, she made some changes to the edges because, “once it was put into numbers, my home and office [were] left out of it, and we restructured it and we were able to correct that.” TrA1278. When asked, “I think what you have indicated to this Court is that you specifically and directly communicated to the drafters of the map that your home and your office were left out of the proposed district and you wanted them to correct that,” she responded “Yes. It was corrected in the map that the delegation submitted. It was not corrected in the map that came out of the legislature.” TrA1278-79 (Johnson). Considering all of the evidence, it appears that Johnson was referring to the delegation map initially leaving out her home and office, not Plan C125, and that she spoke with Smith and Opiela about ensuring that her home and office were in the delegation’s proposed district. Her other testimony indicates that she did not learn that her home and office were not in the Legislature’s map until the map was adopted and that she did not contact Downtown or mapdrawers in the Texas legislature about the omission until after the map was adopted and had become law. Even if she did learn that her home and office were out in Plan C125/C130, her testimony states only that she contacted Smith and Opiela about the issue, not Downton or mapdrawers in the Texas Legislature. 415a CD30. Similarly, the addition of CD36 in southeast Texas did not force major changes to CD9 and CD18 in Harris County. 564. Benchmark CD9 was a diverse district; at the time of redistricting it was 36.7% Black, 42.4% His- panic, 10.4% Anglo, and 11.7% other in terms of total population. Tr1332 (Green); Red-100. When the district was initially drawn it had a clear African- American plurality, but at the time of redistricting Hispanic total population was beginning to exceed the Black population. Tr1332, Tr1337 (Green); TrA1409 (Murray). Green described its communities of interest as including Hiram Clarke and Sunny Side, primarily African-American areas, and the Fondren area, which is heavily Latino. Tr1332 (Green). He said that these communities of interest worked together to elect Sheriff Adrian Garcia, the first Latino sheriff. Tr1333. He also noted the “international district” in Bellaire and the southwest portion of the district, which included diverse Asian populations. Tr1333. 565. Green and Murray testified that benchmark CD9 was close to ideal population, only 34,000 over, was reasonably compact, and was effective, so it could easily have been adjusted for voting rights and practical political traditional principles with modest modification, such as by moving six or so precincts to CD29, which was a little underpopulated (approximately 21,000), next door. Tr1045, TrA1387 (Murray); Joint Expert Ex. E-4 (Murray report) at 28; Tr1336 (Green). Instead, CD9 was substantially and unnecessarily changed, without any consultation with Green. TrA1387- 88 (Murray); Tr1336 (Green). Green’s office was removed, as was the Astrodome, parts of the Medical Center, the veteran’s hospital, and the 90A rail line (that he had been working on funding for). Tr1335-36, 416a Tr1339 (Green); TrA1387 (Murray); D-470.51 In terms of economic engines, Dr. Murray testified that Green was a “net loser.” TrA1388-89 (Murray). CD9 also lost Hiram Clarke, which is African-American, while gaining heavily Latino and upscale areas like Gulfton, Sienna Plantation, and Shadow Creek, which might change the character of the district as an African- American district as the area develops. Tr1336, Tr1339 (Green). As drawn in Plan C185, the total Black population is 38.3% and the total Hispanic population is 38.8%, which could create tension, and the addition of upscale communities (Sienna Plantation, Shadow Creek) will likely produce an influx of Anglos and pose a challenge to maintaining it as an African-American opportunity district. Tr1343-44, Tr1361, Tr1364 (Green); Red-100. 566. Dr. Murray testified that Plan C185 pretty much dismantled and rebuilt CD9, moving a couple hundred thousand people around, taking out the core, including Al Green’s office, and also removing important assets like parts of the Medical Center that he has an important constituent relationship with. This did not follow the principle of starting with existing districts if they do not need much modification. Tr1046,

51 Green testified that the Medical Center provides an opportunity to interact with colleagues and be mutually beneficial to each other, and provides “a kind of prowess that’s recognized in our political order.” Tr1356. The rail line allows a member to work with other members on the rail line. Tr1356-57. Green thought the district was left as a “bedroom community” without prowess. Tr1357. Murray testified that Green gained Reliant Park and a significant portion of the Medical Center (TrA1388), but this testimony is in error, since CD9 lost both Reliant Park and portions of the Medical Center to CD18 (Jackson Lee’s district). D-470. Green also lost Houston Baptist University to CD18. D-501. 417a TrA1392. The new district ignored communities of interest, member constituent relations, and the population dynamics of the district. TrA1392 (Murray). Green’s political base was in the older, inner-city precincts of Harris County, but a lot of new suburban and not-yet-fully-developed areas were added, which creates unpredictable new growth in what had been a mature district, and creates the potential for a tension district or one that loses its character as an African- American district. Tr1046, Tr1050, TrA1392 (Murray); Joint Expert Ex. E-4 (Murray Report) at 28. The new portions of Fort Bend County added in are much less African American and much more Asian and Hispanic, which creates a potentially very different district. Tr1047 (Murray). Enacted CD9 is 38.3% Black and 38.8% Hispanic in terms of total population, giving it a Hispanic plurality. Red-202 Report; Tr1059, TrA1439 (Murray). Dr. Murray opined that the population creates a lot of potential for tension, particularly if there is no other opportunity district for His- panics in the Houston area given the explosive growth. Tr1048. Dr. Murray noted that faster Hispanic population growth in the area increases the possibility that over the decade competition between Hispanics and Blacks in Democratic primaries could threaten the status of CD9 as an African American opportunity district. Tr1440; Joint Expert Ex. E-4 (Murray Report) at 28. He testified that although there is strong evidence of African-American and Hispanic coalition voting, without better opportunities for a growing Latino population as we move through the decade, there will be some real tensions. Tr1061.52

52 Green testified that a “Black and Brown Coalition” exists in Houston. Tr1333. Jackson Lee also testified about a Hispanic- Black-Asian coalition. Tr1521. Murray agreed that, in the 418a 567. As drawn, Green believes CD9 will not continue to perform as an African-American opportunity district as time passes. Tr1364. African Americans and Latinos work well together when one group has a plurality, and will have a “gentle person’s agreement” to back the candidates from the plurality, but when there is not, tension may develop because both groups want representation. Tr1364-67, Tr1346 (Green). Green believes the creation of such tension in CD9 was unnecessary because a Latino opportunity district could have been drawn and instead Latinos were packed into CD9. Tr1345, Tr1373. Green acknowledged that as drawn the BCVAP of CD9 is 47.5% and the HCVAP is only 18.3%, but he believed that, with the predicted growth, the tension would develop. Tr1379. 568. 568. The core of CD18, which was the first district created under the 1965 VRA, included Third Ward, MacGregor, South Park, Acres Home, downtown (Central Business District), Fifth Ward, northeast Houston, and part of the Heights. Tr1509 (Jackson Lee). Jackson Lee noted it was anchored by two high schools in the Third and Fifth Wards, as well as Texas Southern University. Tr1510. Dr. Murray testified that benchmark CD18 was overpopulated by about 24,000, so only minor changes were required, but major changes were made. Tr1051; Joint Expert Ex. E-4 (Murray report) at 29-30. Jackson Lee felt that C185 did not reflect her input or that of her constituents, and did not respect communities of

primary, African Americans tend to vote for African Americans, and Hispanics for Hispanics. Tr1061. Despite African-Americans and Hispanics being “cohesive at a very high level in general elections . . . if [mapdrawers] balance Hispanics and African- Americans about evenly [in a district], you are going to get some tension in Democratic primaries.” TrA1493 (Murray). 419a interest. Tr1511, Tr1520. CD18 lost downtown/the Central Business District, its main economic engine, that had been in CD18 since its creation, as well as part of Third Ward and MacGregor, almost to Jackson Lee’s house. Tr1512 (Jackson Lee); Tr1051, TrA1390 (Murray). 53 Third Ward and MacGregor is a traditional black community that has been in CD18 since 1972 and is split between CD9 and CD18 in Plan C185. TrA1452, TrA1491 (Murray); Joint Expert Ex. E-4 (Murray Report) at 30. Jackson Lee had been working to develop relationships between the downtown business community and constituents in CD18, and removing the economic interests would hurt the district’s residents. Tr1514, Tr1521 (Jackson Lee); Joint Expert Ex. E-4 (Murray Report) at 29-30. Her main office was also removed from her district. Tr1515 (Jackson Lee). 569. Dr. Murray and Jackson Lee noted that areas were added that had no connection to the existing district. Tr1514 (Jackson Lee); TrA1393 (Murray). Dr. Murray testified that there was major unnecessary surgery that negatively impacted CD18 and the member- constituent relationships there. Tr1051 (Murray). Murray and Jackson Lee felt that although it was a performing African-American district as drawn, it had the potential to become a tension district. TrA1394-95 (Murray); Tr1538 (Jackson Lee). And by increasing

53 Dr. Murray characterized downtown as the single most vital economic engine in Harris County. TrA1391. Murray testified that it is important to members to have access to major corpora- tions to work with them on projects in return for creating better opportunities for their constituents, which are usually poor in minority districts. TrA1391-92. Downtown was moved to CD29, Congressman Gene Green’s district (a Latino district), and Murray testified that Gene Green had not lobbied for this. TrA1390, TrA1447-48; D-473. CD18 kept Texas Southern Univer- sity and the University of Houston. TrA1453 (Murray); D-474. 420a Anglos and Asian-Americans (while inner city African- American areas will likely lose black residents), the district is at risk of no longer being an effective African-American district in the next decade. TrA1442 (Murray); Joint Expert Ex. E-4 at 30. 570. In addition to her home, Johnson’s district office was drawn out of her district. TrA686 (Johnson); D-453; D-456. 80.7% of CD30 remained the same between C100 and C185, but areas that were removed included economic engines. TrA709-10 (Johnson); D- 403. She lost the American Airlines Center (home of the Mavericks and Stars), “all of downtown and the arts district, all of the redevelopment that [she] worked very hard to make sure that that core of the city could come back into the greenfield from having brownfields,”54 Love Field airport (for which she had done a lot of work to help expand), and Balch Springs. Tr1282, Tr1296, TrA692 (Johnson); NAACP-7255; D- 449; D-450. She felt the core and the economics of her district, areas that she had worked hard to develop and obtain funding for, had been removed. Tr1276, TrA696 (Johnson). CD30 retained Parkland Hospital,

54 She testified that she got “brownfield” money to turn an old packing house into the American Center and spurred a lot of clean-up and development in the area. TrA696-97. She testified that losing economic areas and downtown hurt her constituents because the business areas, where she would get assistance for projects such as computer centers, had been cut from her district. Tr1297. 55 This exhibit lists the following items that were no longer in her district in Plan C185: Love Field, Downtown, Uptown, Baylor Medical Center, West Village, Scottish Rite Hospital, Old Park- land, Deep Ellum, and Victory Park. NAACP-72. All of these except Love Field are located in and around downtown. Love Field was placed in CD6, and downtown and the other economic engines around downtown were placed in CD32. 421a UT Southwestern Medical Center, and Margaret Hill Hunt Bridge. TrA713 (Johnson); D-445; D-446. With regard to Margaret Hill Hunt Bridge, Johnson noted “there are economic engines, and there are economic engines.” TrA713 (Johnson). She testified that she also got a large federal prison FCI Seagoville 56 that she had not had before, as well as jails, and that was not necessary. Tr1276, Tr1280, Tr1282 (Johnson); Tr1516 (Jackson Lee). She felt that Anglo members got replacements for items they lost, and that she lost more economic engines than other districts did. TrA715, TrA727 (Johnson). 571. CD30 in Plan C185 is 45.6% Black in total population and 40.3% Hispanic in total population. The combined B+H total population is 85.2%. Red-100 Report; Tr1040 (Murray). From the benchmark CD30, BVAP increased from 42.5% to 46.5%, HVAP increased from 34.7% to 35.6% (B+H total population increased from 81.3% to 85.2%). BCVAP increased from 49.8% to 53.1% and HCVAP increased from 19.8% to 20.6%. All Black and Hispanic population metrics increased from the benchmark. Murray testified that CD30 is packed, and this “creates tension - putting so many minorities in a single district that are about equally divided, particularly over a 10-year period,” that will “create some real stressful issues especially since there are no other districts around that minority voters are going to have much of an opportunity to elect anybody of their choice.” Tr1040. Murray also noted that the inclusion of suburban growth areas not likely to support the same candidates as black voters undermines the long-term viability for non-incumbent African-

56 FCI Seagoville appears to actually be located just outside CD30, in CD5. D-457. 422a American candidates to win. Joint Expert Ex. E-4 at 37. 572. Congresswoman Johnson felt that CD30 unnecessarily paired large pockets of minority population who had enough population to represent themselves as a Latino district and an African-American district, and included a large federal prison, which increased the African-American population falsely. Tr1276, 1281. She had concerns about its ability to continue as an effective African-American district. Tr1295. Congress- woman Johnson stated that “[b]y the way CD 30 was drawn, it creates tension among the racial groups in the district because everybody wants representation. They want to be able to identify some of the people that look like them, that think like them, that live with them, to represent them. And when that is totally ignored and almost by design denied, it is a very big problem.” Tr1290. Johnson characterized it as “throw- ing a few crumbs out there . . . for the minorities to fight over.” Tr1302. African-Americans and Hispanics in Dallas vote together in numerous primaries, including sheriff, city council, judge, school board, and sometimes African-American voters supported a Latino and vice-versa. Tr1307, 1314 (Johnson). However, she testified that because they have been deprived of representation, when they see an opportunity to represent themselves with someone who looks like them, that is what they want, and they do not want to be abused by being packed all in one district. Tr1303. 573. Johnson’s district office was removed from CD30 and Al Green’s district office was removed from CD9 in C185. TrA1387 (Murray). Three of Congress- woman Sheila Jackson Lee’s four district offices stayed in her district; but she lost her main district office, 423a which was downtown, along with the whole of downtown to CD29. TrA1463-67 (Murray); D-479. District offices are very important for minority members because they have high demand for constituent services. TrA1411 (Murray). 574. Murray’s 2011 testimony that white members “all kept their offices” was incorrect. TrA1459, TrA1467- 69, TrA1477-78. Congresswoman Johnson testified that she was not aware of any Anglo members losing their district offices, but she had not personally researched the issue. TrA719-20. Anglo congresspersons also lost offices, including Lamar Smith, Joe Barton, Michael McCaul, John Culberson, and Kevin Brady. TrA720; TrA1467-1470, TrA1487; D-489; D- 490; D-491. Mike McCaul and Joe Barton each lost two district offices. Tr1471-76; D-492; D493.57 575. In 2014, Dr. Murray still felt that the loss of district offices by the African-American congresspersons was evidence of discrimination because it reflected their lack of input in the mapdrawing process. TrA1459, TrA1490-91. In all, three of three black members had offices moved while a much smaller percentage of Anglo members had offices moved, and he felt that was “unlikely.” TrA1481-82. Dr. Murray opined that Anglo members’ offices might have been moved due to the creation of new districts, shedding excess population from the districts, or moving into more favorable areas politically, but African-American members’ districts did not need to be redrawn because they were very close to proper population. TrA1482-

57 Although not discussed at trial, Defendants submitted their interrogatory response to show that Ted Poe, Ron Paul, Ruben Hinojosa, Bill Flores, Charlie Gonzalez, Francisco Canseco, Lloyd Doggett, Blake Farenthold, Henry Cuellar, and Gene Green also lost district offices. D-716. 424a 83, TrA1487. He also noted that some of the Anglo members’ offices might have been satellite offices, like Lamar Smith’s in Austin. TrA1485-86. 576. Some Anglo members also lost economic engines. Congressman Joe Barton (CD6) lost AT&T Stadium and Rangers Ballpark, but Johnson felt they were replaced by a larger economic engine, Love Field. TrA715-16 (Johnson); D-447. Texas Instruments’ corporate headquarters was moved from CD3 (Sam Johnson) to CD32 (Pete Sessions). TrA717 (Johnson); D-448. The ExxonMobil corporate headquarters and the University of Dallas were moved from CD32 (Sessions) to CD24 (Marchant). TrA718-19 (Johnson); D-499; D-500. CD7 (Culberson) lost Rice University and parts of the Medical Center (Memorial Hermann, Baylor College of Medicine, UT Health & Science Cen- ter, MD Anderson Cancer Center) to CD18. TrA1444- 46, TrA1457-59 (Murray); D-470; D-475; D-476; D-477; D-467; D-469. CD22 (Olson) lost Hobby Airport to CD29. TrA1454-55; D-481. 577. During redistricting Congresswoman Johnson did not talk to Solomons, Seliger, or Interiano because she worked with Lamar Smith and Opiela as instructed and had previously met with legislators, including members of the House and Senate redistricting committees. TrA705 (Johnson). She believed that Opiela was the person who would be working on the maps and directed her input to him. TrA706 (Johnson). Johnson did not contact anyone with the Texas Legislature between May 31, 2011 and when C185 was enacted. TrA730 (Johnson); Tr1019 (Downton). 578. Al Green was “disappointed” in his lack of meaningful input to the map. Tr1351 (Green). There was “not a lot of opportunity to see things as [the map] developed.” Tr1353 (Green). The map was presented 425a as “take it” (not even “take it or leave it”). Tr1353-54 (Green). Jackson Lee felt that they had no input. Tr1517. Jackson Lee felt that she had spoken to the necessary people and was waiting for someone to send her a proposed map or for further delegation meetings. Tr1519-10. Green and Jackson Lee were not informed of changes to their districts until the public map was first released. TrA1389 (Murray). They tried to work through congressional staff members to restore some of the areas that had been removed, including his district office and the Third Ward in CD18. Tr1352 (Green). Green submitted a proposal through Smith but it did not get done. Tr1353 (Green). Downton recalled either Al Green or Gene Green requesting to have his office put back, but Downton said it was too far from his new district and they “couldn’t get it done.” Tr1019-20. 579. Murray opined that the NAACP Plan C193 restored the traditional cores of CD9 and CD18, respecting communities of interest and returning key economic assets like downtown Houston and the Texas Medical Center, and also reconfigured the populations to keep them as effective African-American opportunity districts throughout the decade. He testified that it also provided an opportunity to create a second Hispanic opportunity district in the Houston area. Tr1052; Joint Expert Ex. E-4 at 37. Green and Jackson Lee also testified that Plan C193 respected and maintained the character of their districts. Tr1340 (Green); Tr1525-1526 (Jackson Lee). Murray opined that Plan C193 provides three opportunity districts in the DFW area, rather than only CD30, and restores CD30’s communities of interest and economic assets. Joint Expert Ex. E-4 at 37. 426a 580. Murray thought Cuellar had input into the map and was treated fairly even though he was a Democrat. TrA1500, 1505-06. Cuellar agreed that he was treated fairly with regard to his district. Tr1325 (Cuellar). Murray opined that this demonstrated that only African-Americans were intentionally ignored or discriminated against. TrA1506. 581. Mapdrawers did not intentionally discriminate against African-American representatives by removing economic engines, offices, or homes from their districts. Downton paid no attention to economic engines, offices, or homes unless notified by affected members. 582. The African-American congressional representatives lacked input into the map, and the changes to the African-American districts were the result of that lack of input. However, the Court does not find that this was the result of discrimination against African Americans specifically. Republican members were in contact with Downton through Opiela or directly and were able to exert some influence in the map. Though Cuellar may have felt he was treated fairly, there is no indication that he was granted specific access to the mapdrawing process that Downton, Green, and Jackson Lee were not. And even if he were, that difference can be explained by location. In the Houston and DFW areas, Downton and mapdrawers were advancing political goals and were unlikely to request input or accommodate requests from any Democrat members. Plan C185 - general findings 583. The benchmark Plan C100 was a court-ordered remedy following remand in LULAC v. Perry, 548 U.S. 399 (2006) for § 2 violations in CD23. Plan C100 had 32 districts. 427a 584. Plan C100 had seven districts (15, 16, 20, 23, 27, 28, 29) with an HVAP above 50% and both total and non-suspense SSVR over 50%. Thus, 21.9% of the districts in Plan C100 contained an HVAP majority and an SSVR majority. 585. Plan C100 had seven districts (15, 16, 20, 23, 27, 28, 29) with an HCVAP above 50% according to the ACS 2005-2009 data. Thus, according to the ACS 2005-2009 survey, 21.9% of the districts in Plan C100 contained an HCVAP majority. 586. Districts 15, 16, 20, and 29 performed 100% (5/5) on Handley’s 2011 Exogenous Minority Effective- ness Index and 100% (10/10) on the OAG 10.58 587. District 23 performed 40% (2/5) on Handley’s 2011 Exogenous Minority Effectiveness Index, 50% (3/6) on Handley’s 2014 Exogenous Minority Effective- ness Index, and 30% (3/10) on the OAG 10. 588. District 27 performed 60% (3/5) on Handley’s 2011 Exogenous Minority Effectiveness Index and 60% (6/10) on the OAG 10.

58 OAG 10 performance is shown on the RED-400 Report. Handley’s 2011 Exogenous Election Index included five top-of- the-ticket statewide elections that included a minority-preferred minority candidate. TrA593 (Handley). She included one election per year between 2002 and 2010 (2010 Lt. Governor; 2008 U.S. Senate; 2006 Lt. Governor; 2004 Court of Criminal Appeals Place 6; and 2002 Governor) to look at elections over the course of a decade to see if patterns emerged and to look at different circumstances. TrA593-94 (Handley); US-686 at 4. In her 2014 rebuttal report concerning CD23, Handley included six elections in her exogenous election index, adding the 2012 Presidential election to her previous five. US-687A. 428a 589. District 28 performed 80% (4/5) on Handley’s 2011 Exogenous Minority Effectiveness Index and 80% (8/10) on the OAG 10. 590. Plan C100 had seven Latino opportunity districts: CD15, CD16, CD20, CD23, CD27, CD28, and CD29. There were no Latino opportunity districts in the DFW area. CD29 was located in Houston/Harris County. The other six Latino opportunity districts were located in South and West Texas. CD23 and CD27 did not perform for minority voters in 2010. 591. Plan C100 had no districts that were above 50% BVAP and had one district (CD30) that was majority-BCVAP using 2005-2009 ACS data. CD30 was located in Dallas County. CD30 was 42.5% BVAP, 49.8% (+/- .8) Black alone CVAP, and 50.3% combined BCVAP (49.8% Black alone + .2 Black+White + .3 Black+American Indian). CD30 performed 100% (5/5) on Handley’s 2011 Exogenous Effective Index and 100% (10/10) on the OAG 10. 592. Districts 9 and 18 were located in Houston/ Harris County and were considered African-American districts. CD9 was 36.3% BVAP, 48.2% Black alone CVAP, 48.4% combined BCVAP, and it performed 100% (5/5) on Handley’s Exogenous Effective Index and 100% (10/10) on the OAG 10. CD18 was 37.9% BVAP, 46.4% Black alone CVAP, 46.7% combined BCVAP, and it performed 100% (5/5) on Handley’s 2011 Exogenous Effective Index and 100% (10/10) on the OAG 10. 593. Plan C100 also had CD25, a district in which minority voters were able to elect their candidate of choice (Lloyd Doggett) with votes from Anglo voters. 594. Plan C185 has 36 districts. 429a 595. Plan C185 has eight districts (15, 16, 20, 23, 28, 29, 34, 35) with HVAP above 50%. Thus, 22.2% of the districts in Plan C185 contain an HVAP majority. 596. Plan C185 has seven districts (15, 16, 20, 23, 28, 29, 34) with both total and non-suspense SSVR above 50%. Thus, 19.4% of the districts in Plan C185 contain an SSVR majority. 597. Plan C185 has eight districts (15, 16, 20, 23, 28, 29, 34, 35) with HCVAP above 50% according to the ACS 2005-2009 survey. Thus, according to the ACS 2005-2009 survey, 22.2% of congressional districts in Plan C185 contain an HCVAP majority. 598. Districts 16, 20, and 29 maintained performance of 100% (5/5) on Handley’s 2011 Exogenous Minority Effectiveness Index and 100% (10/10) on the OAG 10. CD15 dropped from 100% to 80% (4/5) on Handley’s 2011 Exogenous Minority Effectiveness Index and 90% (9/10) on the OAG 10. Districts 15, 16, 20, and 29 are Latino opportunity districts in Plan C185. 599. District 27 dropped from 60% (3/5) to 0% (0/0) on Handley’s 2011 Exogenous Minority Effectiveness Index and 60% (6/10) to 0% (0/10) on the OAG 10. CD27 is no longer a Latino opportunity district in Plan C185. 600. New CD34 performs 100% on both Handley’s index and the OAG 10. New CD35 performed 100% on both indices. 601. District 28 increased from 80% (4/5) to 100% (5/5) on Handley’s 2011 Exogenous Minority Effective- ness Index and from 80% (8/10) to 100% (10/10) on the OAG 10. CD28 is a Latino opportunity district in Plan C185. 430a 602. District 23 performance dropped from 40% (2/5) to 0% (0/5) on Handley’s 2011 Exogenous Minor- ity Effectiveness Index and 0/6 on Handley’s 2014 Exogenous Minority Effectiveness Index and dropped from 30% (3/10) to 10% (1/10) on the OAG 10. 603. Plan C185 has no districts with a BVAP majority and has one district (CD30) with a BCVAP majority using 2005-2009 ACS data. CD30 is 46.5% BVAP, 53.1% Black alone CVAP, and 53.5% combined BCVAP. It performs 100% (5/5) on Handley’s 2011 Exogenous Effectiveness Index and 100% (10/10) on the OAG 10. 604. CD9 is 37.6% BVAP, 47.5% Black alone CVAP, and 47.7% combined BCVAP and performs 100% (5/5) on Handley’s 2011 Exogenous Effectiveness Index and 100% (10/10) on the OAG 10. 605. CD18 is 40.5% BVAP, 48.6% Black alone CVAP, and 48.9% combined BCVAP, and performs 100% (5/5) on Handley’s 2011 Exogenous Effective Index and 100% (10/10) on the OAG 10. 606. In Plan C185, CD25 will no longer elect minority candidates of choice. 607. In 2010, ideal population for the 36 congressional districts was 698,488. 608. In 2010, all of the South Texas Latino opportunity districts were overpopulated.59 CD15 was overpopulated by 88,636. CD16 was overpopulated by 58,939. CD20 was overpopulated by 13,144. CD23 was overpopulated by 149,163. CD27 was overpopulated by 43,505. CD28 was overpopulated by 153,336. The

59 All but four districts (CD13, CD19, CD29, and CD32) in the state were overpopulated. 431a combined excess population (506,723) adds up to 72.5% of a new congressional district. 609. The benchmark congressional districts with the largest increase in absolute numbers of HCVAP between 2000 and 2010 were 23, 28, 16, 15, 21, 27, 10, and 22. TrA964-65 (Ansolabahere); D-439. Five of these districts (23, 28, 16, 15, and 27) were already Hispanic-majority districts. TrA965 (Ansolabahere). These were located in “the envelope” of south Texas. Id. It was possible to create an additional HCVAP- majority district in the envelope without disrupting CD25 as an effective district for minorities. TrA961 (Ansolabahere). 610. The primary persons responsible for drawing and making decisions about the congressional map were Ryan Downton, Burt Solomons, and Kel Seliger. 611. Eric Opiela was working for Congressman Smith and the Republican congressional delegation as their interface with the redistricting committees. TrA1595 (Downton). Opiela and Interiano discussed how to protect Farenthold and Canseco beginning in November 2010. US-75; US-76; TrA343-44 (Interiano). In his November 17 email to Interiano, Opiela proposed the nudge factor, which involved increasing a district’s total Hispanic or HCVAP population while keeping the SSVR and turnout low. In his November 20 email to Lamar Smith (copying Straus’s Chief of Staff Denise Davis), Opiela suggested focusing on CD23 for “enhancement” because protecting Farenthold by taking Nueces County out of the South Texas districts meant the South Texas districts would “have to go North to pick up Anglo voters” in order to avoid a packing claim and they would potentially have to give “Western Bexar to Cuellar to pick up Anglo voters,” 432a which would “severely damage Canseco’s reelect-ability.” US-76. He also noted that they could “play a lot with the large body of people in Bexar [County] to ensure [Canseco’s] minority majority status AND make it still Republican.” Id. Interiano understood Opiela’s nudge factor proposal and tried to obtain the relevant data from the TLC, but TLC did not have not all of the specific data Opiela had wanted. 612. Solomons testified that the HRC tried to solicit input from the various congressional members. TrA1318 (Solomons). Solomons met with MALDEF (Nina Perales) about redistricting. Tr1565 (Solomons). Solomons told the congressional members to come up with proposals to bring to the Legislature. Tr1580 (Solomons). Solomons had conversations about congressional redistricting with Lamar Smith, Mike McCaul, Randy Neugebauer, Pete Olson, Gene Green, Sheila Jackson Lee, Ken Marchant, Kevin Brady (in the hall), and Henry Cuellar (in the hall). Tr1579, TrA1313 (Solomons). The fact that Solomons spoke to congressional members and minority groups did not mean that their input was reflected in the map, however. 613. Solomons testified that Congressman Smith stopped by his office on several occasions (every two or three weeks) during the regular session to update him generally on the delegation’s progress. TrA1271- 72 (Solomons). In early April, Downton received a proposal from Congressman Smith on behalf of the congressional delegation.60 TrA348 (Interiano); TrA1594

60 Smith and Opiela submitted a map on behalf of all the Texas congressional representatives except for Congressman Joe Barton. Dub Maines submitted a map to Downton on behalf of Con- gressman Barton, but Downton did not like the map and thought it would be subject to legal challenge. TrA1605-06, TrA1706 (Downton). Barton had different ideas than Smith had. TrA1297- 433a (Downton); PL-311; D-573. It was a 2-2 plan, meaning that two of the new districts were Republican districts, and two were Democrat/VRA districts (including a minority coalition district in DFW). TrA1705 (Downton); Tr1321 (Cuellar); PL-1114. The 2-2 plan was a compromise plan created with input from the Republicans and Democrats in the Texas congressional delegation. Tr1320 (Cuellar); Tr1277-81 (Johnson). Congressman Smith advocated for this plan throughout April and May. Downton shared the delegations’ proposal with Solomons and Interiano, but not others. TrA1601 (Downton). He did not tell Vice-Chairman of the HRC Mike Villarreal that they had this map. TrA1787 (Downton). On April 28, Solomons denied having seen this map when Rep. Veasey, another member of the HRC, asked him on the House floor about his plans to unveil it or for the House or HRC to consider, and Solomons did not tell Veasey he had received it. D-190 at S847; TrA1313 (Solomons). 614. Downton agreed with the concepts of anchor- ing Farenthold’s district in Nueces County and drawing a new Cameron County-based district in South Texas in the delegation’s map, and he incorporated these concepts into his draft map. TrA1594 (Downton); Downton 8-12-11 depo. at 31. He considered the new Cameron County-based district (CD34) to be a “swap” for CD27, which was no longer a Latino opportunity district. Downton 8-12-11 depo. at 32. Downton rejected other concepts in the delegation’s proposal. TrA1594- 95 (Downton). He rejected the proposed DFW minority district because it was not 50% HCVAP and he

99 (Solomons). Solomons was angry with and would not take calls from Barton because he felt Barton interfered with the House map for Tarrant County and had sued Solomons. TrA1297 (Solomons). 434a concluded that it was not required to be drawn under the VRA. TrA1601 (Downton). 615. In the past, the Texas Legislature had deferred to the congressional leadership and their input Tr1320 (Cuellar); Tr444-45 (Flores). Cuellar thought the Leg- islature would accept the delegations’ proposal since they had a bipartisan deal. Tr1323 (Cuellar). It was unusual how little input members of Congress had into the map. Tr1517 (Jackson Lee). Solomons testified that it was “a fair thing to do” to find out what the congresspersons wanted and did not want, but “at the end, that really wasn’t that important to [him]” because he did not need their votes to pass the map. Tr1633-34. Canseco said “there was a huge disconnect between the Republican Congressional Delegation and the state legislature. They had their own mindset as to what they were going to do.” TrA573. Canseco believed the congressional delegation’s proposal “was a futile exercise.” Id. 616. On May 2, mapdrawers received a proposed plan from Governor Perry, which included an HVAP- majority district in DFW. TrA1602 (Downton); TrA1700; PL-744; D-508.2; D-699. The proposed CD33 in DFW had an HCVAP of 33%, and Downton could not get the district over 50% HCVAP, so he concluded it was not required to be drawn. TrA1604-05 (Downton). Like the delegation’s map, this map was never presented to the House. TrA1603 (Downton). 617. Initially Solomons and Seliger agreed that Seliger would take the lead on the congressional map. TrA340 (Interiano); TrA1596 (Downton); TrA1269 (Solomons). However, both Downton and Doug Davis (from the Senate Committee) were working indepen- dently on draft maps. Downton 8-12-11 depo. at 125- 26. Downton “played around” with a congressional 435a map after the census data came out, but did not focus on drawing the congressional map until the House map was passed. Tr975, TrA1591 (Downton); Downton 8-12-11 depo. at 30, 63. Solomons wanted Downton to be working with the Senate Committee (specifically Doug Davis) on a map so they were not caught “flat footed” if the Governor called a special session. TrA1269, TrA1327 (Solomons). Solomons’ priority was the House map, however, because he did not want it to go to the LRB. TrA1326 (Solomons). 618. Downton was working on the congressional map by May 3. TrA1608 (Downton); D-508. Solomons asked Downton to review the various map proposals and come up with a draft map for his review. TrA1591- 92 (Downton). Downton 8-12-11 depo. at 30-31. Downton counted the number of majority-minority districts in the benchmark (to be sure he kept at least that many for § 5 purposes) and tried to draw majority-minority districts on his own in different places around the state because Solomons had directed that the number needed to go up. Downton 8-12-11 depo. at 31, 63- 64. Downton believed that there were seven Latino opportunity districts (CD15, 16, 20, 23, 27, 28, 29) in the benchmark. Downton 8-12-11 depo. at 70. He looked at where to put the new districts based on population growth in general, and also tested possible VRA districts. Tr904, TrA1598, TrA1763 (Downton); Downton 8-12-11 depo. at 64. He decided that a new district would be required in central/south Texas, and he decided to put that district from Austin to San Antonio, replacing CD25 as a Democrat district.61 He

61 MALDEF submitted two 12-district proposals, Plan C122 and Plan C123. MALDEF’s Plan C122 (MALDEF Proposal A), had eight HCVAP-majority districts (15, 16, 20, 23, 27, 28, 33, 36), two districts between 40-50% BCVAP (CD9 and 18; the map 436a also determined that new districts would be placed in north Texas, south Texas, and around Houston. Tr904, TrA1598, TrA1763-64 (Downton); Downton 8-12-11 depo. at 64. He then had to determine whether the north Texas/DFW district had to be a VRA district, and he turned on racial shading to determine whether he could draw a 50% HCVAP district. He could not draw one, and concluded that the DFW area district did not have to be a minority district. TrA1763-67 (Downton). Because he did not draw a VRA district in DFW, he drew the “lightning bolt” configuration instead. TrA1765 (“If you could do the DFW district, you wouldn’t do the lighting bolt because all of the Republican districts would be so strong if you had a Democratic district in there that you wouldn’t need to pull Democratic population into a district.”). Downton later was alerted that he split Hispanic communities in Fort Worth/Tarrant County and he used racial shading to put them together in the CD26 lightning bolt even though he personally did not think that was required by the VRA and he did not know anything about the communities. TrA1767-68 (Downton). 619. MALDEF and others provided Downton with an analysis of Latino growth statewide. Downton 8-12- 11 depo. at 66-67. He looked at the distribution of the does not include CD30), and two H+BCVAP-majority districts, one in DFW (CD35) and one in Houston (CD29) using both 2005- 2009 ACS data and 2008-2010 ACS data. Joint Map Ex. J-3; D- 546. MALDEF’s C123 (MALDEF Proposal B) created the same number of minority districts without including any part of Travis County. Joint Map Ex. J-4. Downton testified that CD28 in Plan C122 was the basis for the new Austin-San Antonio district in his plan, but he rejected CD35 in DFW because it was not majority- HCVAP and he felt that new CD36 in Houston, which was 50.1% HCVAP, retrogressed CD29 to achieve that number. TrA1593 (Downton). 437a Latino community around the state, including the concentration of Latinos on a county-by-county basis of the percentage of SSVR (by shading). Downton 8-12- 11 depo. at 67. 620. When Downton was working on the congressional map, he was looking at HCVAP, not SSVR, to decide whether a district was a Latino opportunity district. TrA1641 (Downton). Downton felt that the standard for an opportunity district was based on HCVAP, so he considered HCVAP the most important metric. Downton 8-12-11 depo. at 105-106. He did not use SSVR to determine whether any congressional district was a Latino opportunity district. Downton 8- 12-11 depo. at 105. He acknowledged that SSVR was a metric that had been used before (for the House map) and was the only available metric at first. Downton 8- 12-11 depo. at 105. Downton acknowledged that SSVR is generally a little lower than HCVAP. Downton 8-12- 11 depo. at 105. The sole criterion Downton used to determine whether a district was a minority opportunity district was whether it was majority CVAP for a single minority; if it was, then, in his view, it was by definition a minority opportunity district even if it never elected the minority’s candidate of choice. Downton 8-12-11 depo. at 107; Downton 8-31-11 depo. at 25. He could not say whether he was speaking for the HRC on that, and he felt each member of the committee had their own view on what definition to use, though no Republican members of the committee publicly disagreed with him. Downton 8-12-11 depo. at 107-108. He thought Rep. Veasey disagreed with him. Downton 8-12-11 depo. 108. 621. When drawing the maps, Downton looked at the population of the district, population deviation, 438a SSVR, Hispanic population, HVAP, Black total population, BVAP, and different election results. Downton 8-12-11 depo. at 40-41. He used shading in RedAppl for SSVR and for election results. Downton 8-12-11 depo. at 41. 622. Downton knew that turnout and voter cohesion played a role in the effectiveness of a minority district. In his deposition, he noted that the VRA only required an opportunity, and “turnout’s obviously a big factor there.” Downton 8-31-11 depo. at 26-27. Thus, Downton knew that turnout could have a big impact on the effectiveness of a minority district. Downton was also aware that decreasing Latino voter cohesion in a district would allow non-Latino voters to have more of an impact on the outcome of elections in that district. He stated, “[I]t matters whether the Hispanic community is split. If the Latino candidate of choice is preferred by 51 percent of that community, the other candidate by 49, then the rest of the populations can have a large impact on which side wins. I don’t take that to mean the Hispanic community didn’t have an opportunity to elect their candidate of choice. They simply split between two candidates.” Downton 8-31-11 depo. 26-27. 623. In mid-to-late May, Solomons and Seliger decided there would be no DFW minority district. TrA1769, TrA1772 (Downton). Downton advised Solomons that a compact district over 50% HCVAP could not be drawn in the DFW area and that no minority district was therefore required. TrA1598-1600, TrA1770-71 (Downton). Although they knew they could have drawn a district that was less than 50% HCVAP, Republicans in the House and Senate viewed coalition districts and minority districts as Democrat districts and would not draw them unless legally required to do so. TrA277 439a (Seliger); TrA1598-99, TrA1770-71 (Downton), TrA351 (Interiano). Seliger testified that he primarily thought of proposed minority or Hispanic opportunity districts as “democrat districts” and the Legislature would not draw them unless it was required to do so. TrA277. Seliger stated that the focus was to try to create districts that would be represented by Republicans unless § 2 required that they draw minority opportunity districts. TrA270. He stated that the other maps that he received that proposed minority districts “were just designed simply for Democratic seats,” and that proposed maps that created new minority districts were viewed as maps that “would be represented by more Democrats.” TrA287-88, TrA271 (Seliger). Seliger admitted that if a minority legislator proposed a district that was not required under § 2 then he would assume it was a Democrat district and would reject it. TrA293. 624. Once Solomons was advised that only one new VRA district was required, Solomons would only consider a map that increased the net number of Republi- can districts by three. Downton 8-12-11 depo. at 139; TrA1322-24 (Solomons); TrA1599-1600,TrA1770-71 (Downton); Downton 8-31-11 depo. at 115-116; Interiano depo. vol 3 at 68; TrA350 (Interiano). Downton was not convinced that § 2 required a Latino opportunity district to be drawn in south/central Texas, but felt that there was a legal risk if they did not draw one, and it was better to err on the side of caution. TrA1810 (Downton); Downton 8-12-11 depo. at 97. Based on advice from others, Solomons concluded that the central Texas Latino opportunity district was required, and therefore the other three new districts had to be Republican (i.e, not minority) districts and any map had to be a 3-1 map. TrA1604, TrA1810 (Downton). Downton drew a 3-1 map by drawing two new Anglo, 440a Republican districts in the DFW (CD33) and Houston areas (CD36), replacing the existing Democrat district CD25 with CD35, and drawing new CD34. CD34 was a Latino opportunity district drawn to “replace” CD27 as a Latino opportunity district, but it did not replace CD27 as a Republican district because it was already represented by Republican Blake Farenthold. TrA1769- 70, TrA1809 (Downton). Although Downton could have drawn a new central Texas Latino opportunity district without including Travis County or disrupting CD25, he chose this configuration to ensure a 3-1 map. 625. Toward the end of the regular session, Downton and Doug Davis had discussions about their progress on the maps. TrA1596 (Downton); Downton 8-12-11 depo. at 124. They concluded that there was not enough time to get a congressional map passed in the regular session, but they would keep working because there might be a special session. When they learned of the special session, they conferred and concluded that Downton’s map was a better map to work from than what Davis had at that time. Downton 8-12-11 depo. at 125-26; Tr1607 (Solomons). Therefore, in late May, Solomons, Seliger, Downton, and Doug Davis decided that Downton would draw the Congressional map. TrA1596-97 (Downton). 626. Downton was working on a map to present to Solomons in late May. TrA1609, TrA1621 (Downton); D-709. As Downton was drawing districts, he would display population statistics (deviation and numbers of people), as well as the McCain-Obama race statistics (shown as a percentage of people voting for McCain within a district) because most of the Republican congressmen were focused on that race (and preferred at least 55% for McCain in their districts). Tr914, TrA1611 (Downton); Downton 8-31-11 depo. at 95. 441a Downton also turned on partisan shading at the precinct/VTD level using the McCain election. TrA1611-12 (Downton). He also looked at other races. Tr914 (Downton). 627. When working in RedAppl on districts, Downton would pull up election statistics for “what percentage of people within that district voted for candidate X and voted for candidate Y and who would have won if the race had only been in the district,” and he did this for the OAG 10 elections. Downton 8-12-11 depo. at 58, 75. He knew who the Latino preferred candidates were and how their performance was affected as he drew the districts. Id. at 72. He did not look at primaries. Id. 628. Downton testified that neutral redistricting principles such as maintaining cities and communities of interest were honored unless they conflicted with other directives he was given with respect to specific regions. TrA1784, TrA1808 (Downton). Downton got directives from Governor Perry to create Republican districts in Travis County and put Lloyd Doggett in a Republican district (which meant that maintaining Austin and Travis County in a district was not honored and that maintaining the core of district CD25 was not honored). TrA1784-85 (Downton). Solomons assigned Anglo Republican members on the HRC responsibility for their regions, so Downton got directives from Rep. Geren about his DFW area, from Rep. Branch about Dallas/CD32, and Rep. Hunter about Nueces. TrA1785- 86, TrA1804 (Downton). They were all communicating directly with Downton. TrA1785 (Downton). There is no indication that Solomons assigned any minority members of the HRC such responsibilities. 629. Solomons delegated primary responsibility to Downton; it was Downton’s job to put the map together and make recommendations to Solomons. TrA1786 442a (Downton). Solomons trusted his staff was giving him a legal map. TrA1305 (Solomons). Downton took his first map to Chairman Solomons near the end of May. TrA1761 (Downton). 630. Interiano’s role in drawing the congressional map was limited to zeroing out population, and he only spent an hour or two on this. TrA297, TrA342 (Interiano); Interiano 8-26-11 depo. at 65-66. However, he was involved in the mapdrawing process: he received input, communicated with Downton, reviewed maps in progress, and offered his advice and opinions. Interiano 8-9-11 depo. at 23-24. Interiano also went to D.C. in February and communicated with some members of Congress and their staff, including Smith, Poe, Gonzalez, Culberson, Brady, Neugebarger, Paul, and Canseco. Interiano 8-2-11 depo. at 76-77. Interiano met briefly with Congressman Green when he testified at the Houston redistricting hearing; he also met with Con- gresswoman Johnson on the D.C. trip, and he visited with Sheila Jackson Lee’s chief of staff. Interiano 8-26- 11 depo. at 72. However, there is no indication that input from the minority members was incorporated into the map. 631. Opiela visited the redistricting office frequently and spoke to Interiano and Downton frequently as the congressional map was being drawn and revised. TrJ1968-69 (Bruce); Tr1455, TrJ1493, TrA297 (Interiano). Although Opiela did not draw specific lines in the congressional map, Downton spoke with Opiela, reviewed maps drawn by Opiela, and some of Opiela’s ideas were incorporated into the plan. Tr951, TrJ2155, TrA1726-27 (Downton); Downton 8-12-11 depo. at 28; Downton 8-31-11 depo. at 31. 443a 632. No draft congressional plan was released during the regular session. The only hearing on the congressional plan during the regular session was the HRC public hearing on April 7, well before any map was released. TrA895-96 (Dukes). 633. Governor Perry called the special session on May 30. The first public plan (C125), drawn primarily by Downton, was released on May 31. No minority senators had input into the map. D-22 at A-12; TrA257 (Seliger). Attorneys hired by the SRC were not involved in drawing the plan. D-602 at 126. The plan was largely kept secret until it was made public through TLC. TrA1328 (Solomons); TrA1383 (Murray). 634. Downton testified that mapdrawers “were always conscious of the race numbers, and so we would look at them throughout the process before moving forward with the map.” Tr912. He stated that when C125 was released, their analysis was not complete because they wanted to get a map out, “so some things slipped through the cracks initially.” Tr912-13 (Downton). Solomons also stated on the House floor that they had not determined whether any minority-majority districts were performing or protected under the VRA, but that they had maintained minority-majority districts in terms of population. D-601 at 12-13. 635. Notice was given in the House on May 31 or June 1 for a public hearing on June 2 before the HRC and on June 3 before the SRC. Solomons testified that the Senate was going to carry the bill, but they set a hearing in the House as well to provide an opportunity for comment. TrA1330-32. These two hearings, which were set with short notice, were the only opportunities for public comment on the congressional plan. The public had about 48 hours notice before the June 2 hearing, which was the only opportunity for public 444a comment in the House. TrA1345-46 (Solomons). Wit- nesses and members complained that there was insufficient time to consider the plan before the hearing and to make arrangements to come testify. D-601 at 82, 112-13, 116, 178, 224. Minority members also stated their disappointment at the lack of new minority districts in the map. D-601 at 35, 38, 42-43, 112-13, 229. 636. Seliger released Plan C130, a full committee substitute plan, at 6:45 p.m. the night before the Senate hearing on June 3. The public had less than 24 hours to analyze and prepare comments on Plan C130. TrA1348 (Solomons). Many speakers were not aware of the release of Plan C130 and had prepared testimony on Plan C125. Even the three attorneys hired by the SRC stated that they had just received the plan that morning and had not determined whether it complied with the VRA. There was no other opportunity for public comment on the congressional plan. TrA1348 (Solomons). Senators West and Zaffirini complained about their lack of input into the plan and the fact that they had not seen it until it was released. D-602 at 15- 16, 21. 637. On June 6, the Senate debated second and third reading and passed the bill, with all minority members voting against the bill. On June 9, the HRC took up the bill (Plan C144 was made public the day before) in a formal meeting and voted it out the same day (with no public testimony). On June 13, Solomons released Plan C170. It was considered in the House on June 14 during debate, and adopted. On June 15, following minimal further debate, the House passed Plan C185. The Senate concurred in the House amendments and the plan was reported enrolled on June 20. Thus, the congressional plan was made public on May 31 and passed on June 20. 445a 638. The process in which the congressional plan was adopted was rushed. TrA414 (Arrington). It took about 21 days for the congressional plan to pass from its May 31 introduction, which is about two-thirds of the 30-day special session. TrA464 (Arrington). Most of the consideration and amendments were made during the period between May 31 and June 15. In the first phase, the map was drawn by Downton in consultation with various insiders, and that was a closed process. TrA414 (Arrington). After the map was made public, changes were made, but none that would improve minority opportunity. TrA415-16 (Arrington). Dukes testified that the rule requiring all amendments to be prefiled gave members less time to look for solutions or look at data necessary to draw maps. TrA897. She testified that previously they had seven to ten days to see the map. Id. 639. The Governor could have called a second special session if the congressional map did not get passed in the first. TrA1357 (Solomons). There were several special sessions for the 2003 redistricting. Id. 640. Minority legislators felt that they were not given an opportunity to have input into the plan because they did not see it as it was being developed, did not know who was drawing it, and the map was quickly brought to the floor without an opportunity for review through committee hearings. TrA894, TrA901 (Dukes); Tr807 (Turner). Minority members proposed amendments and substitute maps, but they were not adopted. TrA1280-81 (Solomons); TrA416-17 (Arring- ton). There was very little debate on many of the statewide amendments, such as Dukes’ plan C166. TrA899 (Dukes) (less than 30 minutes debate on Plan C166, and most was among minority legislators who supported the plan). Solomons moved to table a number of 446a amendments offered by minority or minority-preferred legislators. TrA1282 (Solomons). The talking points in support of his motions to table were written by his staff (likely Bruce based on information from others), and he did no independent analysis of the amendments. TrA1283-85 (Solomons). He used the same comments for most of the amendments offered by minority members. TrA899-900 (Dukes). 641. Hanna’s role was to provide legal advice, but unlike for the House plan, Hanna was not asked to and did not provide any written analysis of draft congressional plans. TrA1553, TrA1537, TrA1563, TrA1571 (Hanna). He was asked to look at CD23 and the CD12/ CD26 lightning bolt. TrA1569 (Hanna). He knew there was uncertainty about CD23 and whether the plan had § 2 and § 5 issues. TrA1563 (Hanna). He reviewed the proposed congressional plan but could not make a definitive decision about whether it was retrogressive because of the uncertainty on CD23. TrA1565 (Hanna). Hanna was also aware of the argument that the Leg- islature should keep the percentage of minority districts the same when districts are added in order to avoid retrogression, but he was not sure that was a valid application of that rule and did not have a definitive answer on that. TrA1582-83 (Hanna). Hanna felt it was not unusual for him to have a smaller role in the congressional plan because of the involvement of congressmen and their staffs and because there was no County Line Rule issue. TrA1552-54 (Hanna). Hanna felt that the 2011 redistricting cycle was “about the same” as past ones he had been involved in. TrA1554 (Hanna). 642. Solomons still had access to Baker Botts lawyers during congressional redistricting. TrA1267 447a (Solomons). Solomons did not invite minority legislators to seek counsel from Baker Botts. Id. 643. Seliger did not conduct legal reviews of plans himself, but lawyers for the committee or for the House did. TrA279 (Seliger). He would ask them whether districts were required or were protected and “took their advice.” TrA292 (Seliger). Seliger was told by staff that the African-American congressional districts (CD9, 18, 30) were protected and he believed them to be even though they were less than 50% BCVAP. TrA289-90 (Seliger). 644. Plan C185 split 518 VTDs. US-699; TrA409 (Arrington). The precinct splits were overwhelmingly done in the minority districts. TrA410 (Arrington). A statistically significant greater number of splits are located in the minority districts than in majority- Anglo districts. Id. In urban areas, especially in DFW and Houston, where there is much minority growth, the districts are convoluted and there is gerrymandering. TrA1395 (Murray). Outside of urban areas, districts have more traditional shapes, often including whole counties. Id. 645. Downton and Interiano believed that there were seven Latino opportunity districts (16, 23, 20, 28, 15, 27, 29) in Plan C100 and eight Latino opportunity districts (CD15, 16, 20, 23, 28, 29, 34, 35) in Plan C185, an increase of one. Downton 8-12-11 depo. at 70; Interiano 8-26-11 depo. at 93; Tr1455-57 (Interiano). Downton felt that any district with an HCVAP of more than 50% had the opportunity to elect the Latino candidate of choice, if they all vote cohesively. TrA1635- 36 (Downton).62 He felt there were no § 5 problems

62 Interiano did not share this opinion. He believed that the HCVAP percentage was not the sole criterion to determine 448a with C185 because they created one more 50.1%- HCVAP district than existed previously. Although there was disagreement about whether benchmark CD23 was a § 5 district, he felt that it was the same in Plan C185 as it was in the benchmark. And even if turned out that it had been a § 5 ability district and was no longer a § 5 ability district, CD35 was a replacement district, so there was no need for additional § 5 analysis. TrA1636 (Downton). Downton testified that he used the OAG 10 for § 5 compliance but did not feel it was relevant to § 2 given his belief that any district with HCVAP above 50% was by definition an opportunity district regardless of performance. TrA1743-44 (Downton). 646. Dr. Chapa estimated that Plan C185 had eight HCVAP-majority districts (15, 16, 20, 23, 28, 29, 34, and 35), one BCVAP-majority district (30), and two B+HCVAP-majority districts (CD9 is 48% BCVAP; CD18 is 49% BCVAP), for a total of 11 majority- minority-CVAP districts. Joint Expert Ex. E-1 (Chapa Report) at Table 8. 647. Minority congressmembers were very disappointed with Plan C185 and the lack of more minority seats. Tr1289 (Johnson); Tr1318 (Cuellar); Tr1529-30 (Jackson Lee); Tr1355 (Green). 648. Murray opined that minorities control 10 of 36 districts in Plan C185. Tr868, TrA1385. This was a decrease from Plan C100, in which minorities had opportunities in 11 of 32 districts. Tr868, TrA1385 (Murray). Murray opined that 14 to 16 of the 36 districts should be effective for minority voters. Tr1054. whether a district such as CD23 was a Latino opportunity district, and he also considered the election analysis. Tr1457-58 (Interiano). 449a He opined that minorities are therefore worse off in Plan C185 than Plan C100. Tr869. He felt this indicated discrimination given the amount of minority population growth. Tr870. 649. Hispanics’ political influence is greatly diminished by Plan C185. Only 44% of Hispanics of voting age reside in districts where they have a good opportunity to elect candidates of their choice, compared to 88% of Anglos. TrA1402 (Murray). Only 12% of Anglo voters are “stranded” while more than half of Hispanic voters are. TrA1402-03 (Murray). 650. Dr. Engstrom concluded that there were seven districts in Plan C100 that provided Latino voters with reasonable opportunities to elect representatives of their choice: 15, 16, 20, 23, 27, 28, and 20. Dr. Engstrom concluded that there were seven such districts in Plan C185: 15, 16, 20, 28, 29, 34, and 35. He considered Latinos to have a reasonable opportunity in a district when their preferred candidates win the district more often than not, in both the general and primary elections. Tr510-12 (Engstrom); Joint Expert Ex. E-7 (Engstrom Corr. Rebuttal Report) at 28. 651. Arrington opined that in spite of the enormous growth in HCVAP and Hispanic population in Texas, the congressional plan created no additional districts in which Hispanics could elect candidates of their choice, even though the number of seats expanded by four, because CD23 and CD27 were eliminated as opportunity districts. TrA394. 652. Dr. Ansolabahere concluded that Plan C185 falls short of what might reasonably have been expected for Hispanic and Black representation following the 2010 Census and reapportionment. Joint Expert Ex. E-15 at 5. Although Anglos are not a majority of 450a Texas’s population, they are majorities of the VAP in 64% of districts and of the eligible electorate in 69% of districts. Hispanics are 37% of the population but majorities in just 22% of districts. African Americans are 12% of the population but are majorities in no districts and pluralities in 8% of districts. Id. at 5-6. Plan C185 does not increase minority representation even though minorities accounted for most of the State’s growth. The White population grew by less than 500,000 persons, but Plan C185 creates three additional districts in which White non-Hispanics are a majority of the VAP. Id. at 6. Plan C185 divides racial groups such that almost all White non-Hispanics (88%) are in majority-White districts, but most Blacks and Hispanics of voting age are stranded in districts in which they are the minority. No Blacks are in Black- majority districts and only 44% of Hispanics are in HVAP-majority districts. Id. In the five most populous counties, Blacks and Hispanics receive even less representation than one would expect from their populations. Dallas and Harris Counties are notable. Hispan- ics have 40% of the population in those two counties, but only one of the twelve districts contains a majority HVAP. White non-Hispanics are 30% of the population but receive the vast majority of seats. Id. at 7. The Legislature could have chosen plans that would have provided greater minority representation. Id. 653. Plan C121, “the Fair Texas plan” introduced by Rep. Veasey during the session, has eight HCVAP- majority districts (15, 16, 20, 23, 27, 28, 29, 33), three African-American districts (between 45 and 50% BCVAP) (9, 18, 30), and two H+BCVAP-majority coalition districts (34, 35) using 2005-2009 ACS data. Joint Map Ex. J-2. Using 2008-2010 ACS data, 50.1% BCVAP is within the margin of error for CD9 and CD18, but no additional districts become majority- 451a minority CVAP. D-545. Travis County is split between only CD25 and CD23, with the minority population joined in CD23 with Bexar County. Most of Nueces County is joined with Cameron County in CD27, while a small portion is joined with CD14 to the northeast. Lichtman testified that Plan C121 has 14 minority- majority districts, of which 13 are reasonable opportunity districts (CD22 was 50.9% combined minority population but was not effective for minority voters). Tr1238; Joint Expert Ex. E-3 (Lichtman Report) at 15-20. Comparatively, Lichtman noted that Plan C185 has 13 districts with voting age majority minority populations, but in three of those districts (CD6, CD23, CD27) Anglo bloc voting defeats minority-preferred candidates such that Plan C185 has only ten reasonable opportunity districts (CD23 is not one). Tr1237; Joint Expert Ex. E-3 at 15. Lichtman therefore concluded that Plan C121 has three more effective majority-minority districts (CD33, CD34, and CD35) than Plan C185. Joint Expert Ex. E-3 at 19-20. Lichtman opined that one reason for the lower number of effective majority-minority districts in Plan C185 was the “excessive packing of minorities into other districts,” noting that CD9, CD16, CD29, and CD30 are more than 85% majority-minority (compared to only one such district in Plan C121). Joint Expert Ex. E-3 at 20 n.2. 654. Plan C163, offered by Rep. Martinez Fischer, has eight HCVAP-majority districts (15, 16, 20, 23, 27, 28, 29, 33), three districts with BCVAP between 40 and 50% (9, 18, 30), two H+BCVAP-majority districts (5, 35), and one H+B+Asian CVAP-majority district (36) using 2005-2009 ACS data. Joint Map Ex. J-5. Using 2008-2010 ACS data, CD30 is 50.4% BCVAP and CD36 is B+HCVAP-majority. D-562. Travis County is split between only two districts, CD25 and CD34, 452a and is not joined with Bexar County. Nueces County population is placed 28% into CD15, 70% with CD27 going to Cameron County, and 3% into CD14 to the northeast. Dr. Chapa estimated that Plan C163 has eight HCVAP-majority districts (15, 16, 20, 23, 27, 28, 29, and 33), one BCVAP-majority district (30), and five H+BCVAP-majority districts. Joint Expert Ex. E-1 (Chapa report) Table 8. 655. Plan C164, offered by Rep. Martinez Fischer, has eight HCVAP-majority districts (15, 16, 20, 23, 28, 29, 33, 34), three districts between 40 and 50% BCVAP (9, 18, 30), two B+HCVAP-majority districts (5, 12), and one B+H+Asian CVAP-majority district (36) using 2005-2009 ACS data. Joint Map Ex J-6. Using 2008- 2010 ACS data, CD30 is 50.4% BCVAP and CD36 is H+BCVAP majority. D-563. Travis County is split between only two districts (CD10 and CD25) and is not joined with Bexar County. Nueces County population is placed 27% into CD27 going to the northeast and 73% into CD33 joined with Cameron County. Dr. Chapa estimated that Plan C164 has eight HCVAP- majority districts (15, 16, 20, 23, 28, 29, 33 and 34), one BCVAP-majority district (30) and five B+HCVAP- majority districts. Joint Expert Ex. E-1 (Chapa report) Table 8. 656. Plan C166, the Dukes Plan, has seven HCVAP- majority districts (15, 16, 20, 23, 27, 28, 33), three over-40% but less than 50% BCVAP African-American districts (9, 18, 30), and three H+BCVAP-majority coalition districts (29, 35, 36) using 2005-2009 ACS data. Joint Map Ex. J-7. Using 2008-2012 ACS data, CD30 becomes majority BCVAP (50.8%) though no additional HCVAP-majority districts are created (CD36 in Houston becomes 49.1 (+/-.9)%). D-565. It maintains CD25 as a Travis County-based district. All of Nueces 453a County is kept in Latino opportunity districts CD15 and CD27. Dr. Ansolabahere concluded that Plan C166 has fourteen districts in which minorities have a reasonable opportunity to elect their candidates of choice: 9, 15, 16, 18, 20, 23, 25, 27, 28, 29, 30, 33, 35, and 36. Rod-908 (Oct. 24, 2011 Report). He notes that 15, 16, 20, 23, 27, 28, 29, 33, 35, and 36 are majority- HVAP; 9, 18, and 30 are majority-BVAP, and CD25 would be an Anglo-plurality crossover district. Rod- 908. Plan C166 has more split precincts (536) than Plan C185. TrA411, TrA434 (Arrington). 657. Professor Kousser and Dr. Arrington concluded that other plans created more minority opportunity districts than did Plan C185, and that additional minority opportunity districts could be created in DFW, Houston, and South Texas. Joint Expert Ex. E-2 (Kousser report) at 110; TrA406-07 (Arrington). Dr. Arrington cited Plan C163 and Plan C166 as examples. TrA407. He opined that they satisfied traditional redistricting principles at least as well as Plan C185 such that following traditional redistricting principles would not provide a rationale for the alleged inability to draw more minority districts in Plan C185. TrA411. The plans are similar in population deviation and compactness. TrA433 (Arrington). Neither Plan C163 nor Plan C166 has more HCVAP-majority districts than Plan C185. TrA431 (Arrington); D-671. Dr. Arrington did not base his assessments of opportunity only on HCVAP numbers because demographics alone are a misleading indicator of whether a district will actually provide an opportunity. TrA432 (Arrington). All three plans also have three African-American districts. TrA433-34 (Arrington). 658. Most plans offered by Plaintiffs and by minority members and groups during the session reviewed 454a by Professor Kousser created eight HCVAP-majority districts (C121, C131, C163, C164, C165, C187), the same number as in the enacted plan. See Joint Expert Ex. E-2 (Kousser report) at 111 Table 22. However, those plans all created more B+HCVAP-majority districts (13 or 14) than did Plan C185 (which had 11), and all but C121 and C131 created more HVAP- majority and B+HVAP-majority districts than Plan C185. Id. 659. Plan C187 has eight HCVAP-majority districts (15, 16, 20, 23, 28, 29, 33, 34), three districts between 48 and 50% BCVAP (9, 18, 30), and three H+BCVAP- majority districts (5, 12, 36) using 2005-2009 ACS data. Joint Map Ex. J-9. This plan preserves CD25 and puts 73% of Nueces County’s population in CD33 joined with Cameron County. Dr. Chapa estimated that Plan C187 had eight HCVAP-majority districts (15, 16, 20, 23, 28, 29, 33, and 34), one BCVAP-majority district (30) (CD9 is 49.7%), and five H+BCVAP- majority districts, for a total of 14. Joint Expert Ex. E- 1 (Chapa report) at Table 8. 660. MALC Plan C188 (MALC-172) has nine HCVAP-majority districts (10, 15, 16, 20, 23, 28, 29, 33, 34), three districts between 40 and 50% BCVAP (9, 18, 30), two H+BCVAP-majority coalition districts (12, 35), and one H+B+Asian CVAP-majority coalition district (36) in Houston using 2005-2009 ACS data. Joint Map Ex. J-10. Using 2008-2010 ACS data, CD36 is H+BCVAP-majority (51.4%). In this plan, Travis County is divided among five districts, with the eastern portion joined into three Latino opportunity districts 10, 28, and 34. 27% of Nueces County’s population is joined with CD14 to the northeast, and the remaining 73% is joined with CD33 connected to Cameron County. Eight HCVAP-majority districts are placed 455a in south and central Texas, and CD29 is the ninth HCVAP-majority district, located in Houston. The African-American districts 9, 18, and 30 are maintained. New coalition district CD36 is placed in the Houston area, and new coalition districts CD12 and CD35 are placed in DFW. Dr. Chapa estimated that Plan C188 has nine HCVAP-majority districts (10, 15, 16, 20, 23, 28, 29, 33, 34), zero BCVAP-majority districts, and six B+HCVAP-majority districts, for a total of 15 minority districts. Joint Expert Ex. E-1 Table 8. Plan C188 has more HCVAP-majority districts than C185. Tr303 (Kousser ). But CD10 goes from the Mexican border (Hidalgo County) to Travis County and CD28 is similar. Tr304 (Kousser). 661. The Task Force Plan C190 has nine HCVAP- majority districts (6, 15, 16, 20, 23, 28, 34, 35, 36), two districts between 40 and 50% BCVAP (9, 18), one BCVAP-majority district (30), and one B+HCVAP- majority district (29) using 2005-2009 ACS data. Dr. Engstrom concluded that the Task Force’s Plan C190 contains nine Latino opportunity districts: 6, 15, 16, 20, 23, 28, 34, 35, and 36 (utilizing his definition of opportunity, which is that Latinos have a reasonable opportunity when their preferred candidates win more often than not). Tr515; Joint Expert Ex. E-7 (Engstrom Corr. Rebuttal Report) at 28. Seven of the districts are in South Texas, including CD35, which runs from San Antonio to Austin, and is identical to CD35 in Plan C185, and CD34, which connects most of Nueces County to Cameron County (the rest of Nueces County is combined with CD27 to the northeast). CD6 is in DFW and has an HCVAP of 50.4%. CD36 is in Houston and has an HCVAP of 50.1%. CD29 in Houston is a B+HCVAP-majority district that the Task Force contends would be an African-American and Latino coalition district. 456a 662. Plan C192 has eight HCVAP-majority districts (15, 16, 20, 23, 27, 28, 29, 33), three districts between 45 and 50% BCVAP (9, 18, 30), and two H+BCVAP- majority districts (34, 35). It also maintains CD25 as a Travis County-based district, and new HCVAP- majority CD33 is placed in central Texas around Bexar and Guadalupe Counties without including Travis County. 27% of Nueces County’s population is placed in CD14 to the northeast, and the remaining 73% is in CD27 joined with Cameron County. Two new coalition districts (CD34 and CD35) are placed in DFW. 663. The NAACP’s Plan C201 is a 14-district demonstration plan. Joint Map Ex. J-40. It has seven HCVAP-majority districts (15, 16, 20, 23, 27, 29, 33), three African-American districts between 40 and 50% BCVAP (9, 18, 30), and three B+HCVAP-majority districts (28, 34, 35). New CD34 and CD35 are located in DFW, and CD28 is a new central Texas district anchored in Bexar County (it has 48.4% HCVAP). It also maintains CD25 as a Travis County-based district. Nueces County is kept in Latino opportunity districts CD15 and CD27. 664. Plan C262 has 9 HCVAP-majority districts (15, 16, 20, 23, 27, 28, 29, 34, 35), three African-American districts (9, 18, 30, all between 49 and 50% BCVAP), and two B+HCVAP-majority coalition districts (6 and 33) in DFW using 2008-2012 ACS data. MALC-173 (docket no. 1259); LULAC-12. This plan keeps all of Nueces County in Latino opportunity districts CD27 and CD34 and places eight HCVAP-majority districts in south and central Texas, with the ninth (29) remaining in Houston. Although Travis County is divided among four districts, CD25 is anchored there, and the new CD35 Latino opportunity district in central Texas does not come into Travis County. 457a General Fact Findings 665. The overall statewide growth rate of Texas between 2000 and 2010 was 20.6%. The statewide growth rate for Hispanics was 41.85%. The African- American growth rate was 22.1%. The Asian growth rate was 71.1%. The Anglo population growth rate was 4.2%. 666. Approximately 89% of the 4.2 million population growth in Texas between 2000 and 2010 was attributable to minorities/non-Anglos, and 65% of the population growth (2,791,255 persons) was attributable to Hispanics. TrA1819 (Alford). Mapdrawers and legislators were aware of the extensive minority population growth in Texas between 2000 and 2010. TrA241 (Seliger). However, neither the House map nor the congressional map reflected that growth in terms of the number of minority opportunity districts or minority ability to elect districts. 667. HVAP and BVAP data is produced by the Census Bureau in the decennial enumeration. CVAP generally, and HCVAP in particular, is now produced on the American Community Survey. Tr1669 (Rives). VAP numbers are based on the census and reflect actual population counts, while CVAP data are estimates. ACS is the only source of citizenship information from the Census Bureau and is commonly used by people in the field. Ansolabahere 8-22-11 depo. (Ex. J-44) at 12. 668. The Legislature and mapdrawers knew that CVAP data was needed for compliance with § 2 of the VRA. D-578 at part 2, page 23 (Sept. 1, 2010 hearing, Hanna testimony); D-590 at 67 (March 1, 2011 hearing, Archer testimony). HCVAP data was available (based on 2005-2009 ACS data and the DOJ Special 458a Tabulation) to mapdrawers and legislators by April 21 at the latest, before the House map went to the floor. However, the 2005-2009 ACS citizenship data/DOJ Special Tabulation used by the Texas Legislature to redistrict in 2011 underestimated the HCVAP in 2010/2011. The Legislature knew that the available CVAP data had accuracy issues and had larger margins of error than with the decennial census long form, especially for smaller geographic areas. D-578 at part p. 21-23 (Sept. 1, 2010 Senate hearing, Hanna testimony); D-590 at 71 (March 1, 2011 HRC hearing, Hanna testimony). Mapdrawers used the CVAP data provided by TLC and assumed that TLC had ensured its accuracy. Tr1488-89 (Interiano). Downton would not consider drawing any new opportunity district that did not meet a 50% CVAP threshold using the TLC data. 669. Although the most recent CVAP data available to the Legislature was the 2005-2009 ACS data and the DOJ Special Tabulation, the Legislature could have hired experts or utilized the State Data Center to perform extrapolations to estimate 2010 CVAP. TrJ913-17, TrJ923 (Fairfax); D-578 at 18. Dr. Chapa opined that the ACS and DOJ special tabulation should undergo additional analysis in order to be useful and reliable for redistricting. Joint Expert Ex. E-1 at 12. He developed a method to produce a conservative estimate of CVAP populations that involved calculating a citizenship ratio using ACS data and then applying that ratio to the 2010 census data. Id. Chapa’s method produced HCVAP estimates that generally were slightly higher than the TLC’s. Id. at 14. 670. RedAppl displayed only non-suspense SSVR, but reports from TLC displayed total SSVR. TrJ2139, TrJ2146, TrJ2070 (Downton). SSVR data is derived 459a from the Secretary of State’s website, and does not lag. Tr1668-1669 (Rives). However, SSVR data has some accuracy issues, and the Legislature was aware of those issues. D-590 at 75-77 (March 1, 2011 HRC hearing, Archer testimony). 671. Mapdrawers defended the lack of new minority opportunity districts on the basis that SSVR and CVAP growth were not as high as minority total population growth. TrJ1534 (Interiano); TrA1820-22 (Alford). Dr. Alford acknowledged that there was substantial HCVAP growth, but stated that it did not dwarf the other groups as total population did. TrA1820-21 (Alford); PL-631. Data from the 2005- 2009 ACS survey and DOJ special tabulation indicated that between 2000 and 2009, HCVAP increased by about 702,000 persons, BCVAP increased by about 274,000 persons, Anglo CVAP increased by about 487,000 persons, and “others” increased by about 134,000 persons. PL-631; Joint Expert Ex. E-9 (Gonzalez-Baker report). Thus, under that data, minorities accounted for almost 70% of the CVAP growth, while Anglos accounted for 30%. Hispanics alone were 44% of the CVAP growth. 672. Using updated data, Dr. Ansolabahere also found that minorities accounted for most of the CVAP growth from 2000 to 2010. TrA935 (discussing his report, in which he found that Texas added 2.6 million citizens of voting age between 2000 and 2010, of which 64.4% were Hispanic or Black); Rod-912 at 9-10. Using the average of the 2008-2012 ACS data63 and the long- form census data, Dr. Ansolabahere concluded that

63 Unlike the 2005-2009 ACS survey data, the 2008-2012 ACS data aligns very closely with the 2010 Census enumeration. TrA933 (Ansolabahere). 460a HCVAP increased by 1.24 million persons (from 22.3% to 26.5% of statewide CVAP), BCVAP increased by 432,000 persons (from 12.3% to 13% of statewide CVAP), and Anglo CVAP increased by 660,000 (a decrease from 62.5% to 56.4% of statewide CVAP). TrA934-35 (Ansolabahere); Rod-912.64 He also found that HCVAP increased everywhere in the state and that Anglo CVAP as a percent of the population decreased in every congressional district. See also TrA1407 (Murray) (looking at 2008-2012 ACS data, the HCVAP% has increased in every congressional district). Even though minorities accounted for almost two-thirds of the CVAP growth and the Anglo share of CVAP statewide declined, Anglos were given control of three of the four new districts in the congressional map, and the House map failed to recognize the minority CVAP growth. 673. Dr. Ansolabahere also found that Hispanic and Black CVAP growth outpaced Anglo growth in Harris, Dallas, Tarrant, Bexar, and El Paso Counties. TrA935 (Ansolabahere). However, while these counties experienced HCVAP growth, they did not necessarily become HCVAP-majority. Bexar County became majority HCVAP, but Tarrant County remained a little over 15%, Dallas County a little over 20%, and Harris County a little over 25%. TrA1828-29 (Alford). 674. In Travis County, Anglo CVAP growth roughly kept up with minority growth. TrA936 (Ansolabahere);

64 In 2011, Dr. Ansolabahere had estimated statewide HCVAP in 2010 to be 25.8% and BCVAP to be 12.7% by multiplying their CVAP percentage by their total population in the 2010 enumeration. Joint Expert Ex. E-15 (Ansolabahere Aug. 30, 2011 rebuttal report) at 11, 17. These estimates were higher than the 2005- 2009 ACS data and indicated that the 2005-2009 ACS data overestimated the Anglo CVAP and underestimated the BCVAP and HCVAP. Id. at 12. 461a TrA1894-95 (Alford). Dr. Ansolabahere found that benchmark CD25 had the lowest percent decline in Anglo CVAP of all the congressional districts. TrA937- 38; Tr1894 (Alford). Dr. Ansolabahere concluded that there was sufficient HCVAP population growth in the existing south and southwest Texas districts to create a new Latino opportunity district in the congressional plan without affecting CD25 as a district in which minority voters could elect their preferred candidates. 675. Dr. Alford acknowledged that there was an increase in HCVAP throughout the state. TrA1829. He also found that HCVAP increased in all congressional districts. TrA1832. But he opined that it was difficult to create new congressional Latino opportunity districts (as defined by >50% HCVAP) because HCVAP growth either occurred in areas that were already majority HCVAP or was dispersed in other areas such that few majority-HCVAP areas emerged that would be “an obvious target area for drawing new Hispanic districts.” TrA1827, TrA1830 (Alford). In 2000, counties in south and southwest Texas were majority CVAP, and that remained true in 2010, while counties in other areas increased in HCVAP% but mostly remained below 50%. TrA1827 (Alford); D-230; D-231. Dr. Alford agreed with Dr. Ansolabahere’s conclusions that in “the envelope,” Anglo CVAP decreased and HCVAP increased, and that the concentrated HCVAP growth occurred in the southwest Texas districts. TrA937 (Ansolabahere); TrA1837-39 (Alford). Hispan- ics account for almost all of the CVAP growth in the congressional districts in the envelope (15, 16, 20, 23, 27, and 28). TrA939 (Ansolabahere); TrA1827, TrA1832 (Alford). The top eleven districts in terms of HCVAP growth (23, 28, 16, 15, 21, 27, 10, 22, 31, 29, 20) include all seven of the benchmark Hispanic districts. TrA1833 (Alford); TrA1407 (Murray). Dr. Alford noted that in 462a CD10 and CD21, which were not Hispanic-majority districts, there was substantial growth, but it started at a low level and remained at a relatively low level. TrA1833. Only CD29 crossed the 50% threshold. TrA1832 (Alford). Dr. Alford felt that the lack of many proposals from the various plaintiffs creating more than eight HCVAP-majority districts was strong evidence that it was difficult to do so. 676. In 2010, the Hispanic share of the total state population was 37.6% and HVAP was 33.6%. The HCVAP was approximately 26.4% or 26.5%. D-423; D- 39; TrA935 (Ansolabahere); TrA1822 (Alford). 677. In 2010, the African-American share of the total state population was 11.8% and BVAP was 11.5%. The BCVAP was approximately 12.9% or 13%. D-39; TrA935 (Ansolabahere). 678. There is a strong correlation between partisanship and ethnicity in current Texas politics. TrJ170 (Arrington); TrJ1858-59 (Alford). African Americans, Latinos, and to a lesser extent Asians, have tended to identify with the Democratic party in recent years, while more Anglos have shifted to the Republican party, especially in rural areas. Joint Expert Ex. E-4 (Murray report) at 12. 679. Downton was aware that in Texas there is a different party preference associated with race. TrA1801 (Downton). He knew that Anglo voters generally prefer the Republican party statewide, and African- American and Latino voters statewide usually prefer the Democratic party. TrA1801-1802 (Downton). He knew that one of the reasons the Republican congressional delegation proposed the minority coalition district in DFW was because it protected all the incumbent Republican congressmen by taking Democrat 463a (minority) votes and concentrating them in a new district. TrA1606 (Downton). 680. Latinos in Texas are willing to vote for both Democrats and Republicans, and there are Latino Republicans in Texas. Tr1656 (Gonzales); Tr571 (Jimenez); TrJ694-95 (Rodriguez). Seliger testified that Republicans have not done a very good job of reaching out to Latinos and some party positions on immigration do not attract Latinos. TrA233-34. Lay witness Alex Jimenez agreed that if candidates are anti-immigration they will not get support from Latinos, and that is the perception about the Republi- can party. Tr577. 681. African-American and Hispanic voters have been more aligned in recent years due to shared historic past discrimination, similar economic issues, lack of Republican vote courting in recent years, and the rise of the Tea Party. Joint Expert Ex. E-4 (Murray report) at 21-22. Professor Bill Piatt also opined that African Americans and Latinos have common interests in education matters, including funding, curricu- lum, and graduation rates. Joint Expert Ex. E-14 at 3. They also share positions on immigration, economic development, employment conditions, housing discrimination, and business opportunities. Id. Rep. Thompson also testified that African Americans and Latinos have many common and overlapping concerns, such as education, minimum wage, healthcare, and police brutal- ity. TrJ1260. 682. Dr. Murray stated that the trend across the country is “a sharp uptick in polarization,” particularly in the South. TrA1398. The significant Republi- can Hispanic vote has dropped sharply in the last five or six years. TrA1399 (Murray). Murray testified that the Democrats have become the party of default for 464a African-American and Hispanic voters, and there has been increased cohesion between them. TrA1399- 1400. 683. Texas has open primaries. Joint Expert Ex. E- 4 (Murray report) at 18. There is almost no African- American voter participation in the Republican primaries, and very little Hispanic participation. Joint Expert Ex. E-4 (Murray report) at 18. It is difficult to measure racially polarized voting in Republican primaries because of such low minority voter participation. Id. at 19. A Hispanic candidate would have a hard time prevailing in an overwhelmingly Republican district because there are few Hispanic voters in those primaries. Tr1062 (Murray). 684. Using voter files and looking at 2008 and 2010, Dr. Lichtman found that Latino and African-American voters in Texas are cohesive in that both voter groups overwhelmingly choose to participate in Democratic rather than Republican primaries. Joint Expert Ex. E- 3 at 1, 4-6. 685. Dr. Engstrom’s original report (Joint Expert Ex. E-7) looked at primary elections for a variety of statewide offices in 2008 and 2010 in Bexar, Dallas, El Paso, Harris, Nueces, Tarrant, and Travis Counties, and a 52-county area in South Texas involving a Latino and non-Latino candidate. The 2010 primary elections included the Democratic primaries for Lt. Governor (Chavez-Thompson) and Land Commissioner (Uribe) and the Republican primaries for Governor (Medina) and Railroad Commissioner (Carillo). The 2008 primary elections included the Democratic primaries for U.S. Senate (Noriega), and Supreme Court Places 7 (Cruz) and 8 (Yanez) (there were no applicable Republican primaries). Using King’s ecological infer- 465a ence analysis, Engstrom conducted a bivariate analysis (Latino and non-Latino) for each area, and a multivariate analysis (Latinos, African Americans, and others) for the five largest counties (Bexar, Dallas, Harris, Tarrant, and Travis). He found that Latino voters were highly cohesive in their support of Latino candidates in the Democratic primary. 65 In the Democratic primaries, non-Latinos, including African Americans, did not consistently share the candidate preferences of Latino voters. African-American voters shared support for two of five primary candidates preferred by Latinos in Harris County, one of the five in Bexar County, Dallas County, and Tarrant County, and none of the five in Travis County. The bivariate analysis showed non-Latinos did not share support for any Latino-preferred Latino candidate in the Demo- cratic primaries in the South Texas counties. In the 2010 Republican primary for Railroad Commissioner, Latinos generally supported the Latino candidate Carillo but non-Latinos did not (except in Harris County where a majority of African-American voters did support Carillo (54.1%), but other voters did not (42.8%) and in Travis County, where in the bivariate analysis Latinos supported Carillo but a majority of non- Latinos did not and in the multivariate analysis, Latino voters did not support Carillo (46%) while African-American voters did support him (68.9%) and other voters did not (47.1%)). Latino and African- American voters were generally cohesive in their lack of support for Medina in the 2010 Republican primary

65 All five Latino Democratic Primary candidates received strong Latino support in all areas except Travis County, where four of the five (not Chavez-Thompson) received strong Latino support. 466a for Governor (except in Travis County, where African- American voters supported him but Latinos did not). 686. In his (corrected) Rebuttal Report (Joint Expert Ex. E-7; docket no. 307-1), Dr. Engstrom added the 2006 Primary election and runoff election for Lt. Governor. He also made conclusions about racially polarized voting, finding the presence of racially polarized voting in primary elections in El Paso County, Nueces County, South Texas, Dallas County, Harris County, Tarrant County, and Travis County. He again found that Latino voters were very cohesive in their preferences for Latino candidates in Democratic primaries. He found that African-American support for Latino-preferred candidates was not typically present in primaries and, in fact, African-American voters had “a distinct tendency to vote for candidates competing with the candidates preferred by Latinos in primary elections.” Other voters (primarily Anglos) usually cast a majority of their votes for the opponents of Latino- preferred candidates in primary elections as well. 687. Dr. Alford concluded that African-American and Hispanic voters are not generally cohesive in the Democratic primary, and that the polarization is based on the ethnicity of the candidates such that a “Black candidate will not typically be the candidate of choice of Hispanic voters in the Democrat primary” and “the Hispanic candidate will typically not be the candidate of choice of Black voters in the Democrat primary.” TrJ1859-60 (Alford). Professor Kousser also found that Latinos and African Americans were not cohesive in the Democratic primary. Tr265. Kousser found that Latinos vote for Latino candidates in the Democratic primaries, but African-Americans did not tend to support Latino candidates in the 2010 Demo- cratic primaries. Joint Expert Ex. E-2 at 73. 467a 688. Statewide, in general elections, Anglo voters tend to vote for the Republican candidate, and African- American and Hispanic voters tend to vote for the Democrat candidate, regardless of the race of the candidates and even when the Republican candidate has a Spanish surname. Tr267 (Kousser); TrJ1858-59 (Alford); Engstrom Rebuttal report at 24-25, Tables 1- 8; Alford Rebuttal Report (E-17) at 16, Table 1; Joint Expert Ex. E-2 (Kousser report) at 26, 36; Tr1260 (Lichtman) (except in Travis County); Joint Expert Ex. E-3 (Lichtman report) at 1. This correlation is so strong that Dr. Alford assessed whether districts were minority opportunity districts by looking at Demo- cratic results/wins (noting that in Texas, minority candidates of choice means Democrats). Tr1866 (Alford). 689. Given this political landscape, it is difficult to differentiate an intent to affect Democrats from an intent to affect minority voters. TrJ169-70 (Arrington). Making minorities worse off will likely make Demo- crats worse off, and vice versa. Tr312 (Kousser). In addition, the political polarization along racial lines means that partisan gerrymandering results in Anglos maintaining dominance that is not warranted by the declining Anglo population share in Texas. Joint Expert Ex. E-4 (Murray Report) at 38. 690. Regardless of methodology, all the experts found that general election and primary election voting in Texas is highly polarized along racial-ethnic lines. During trial, the State conceded that racially polarized voting exists in all areas of the State of Texas except Nueces and Kleberg Counties. 691. Professor Morgan Kousser found that voting in recent Texas elections has been racially polarized between Latinos and non-Latinos. Joint Expert Ex. E-2 at 26, 30 (“the general election of 2010 in Texas 468a was markedly racially polarized”). He found similarly strong levels of racial polarization in the three 2010 statewide elections involving Spanish-surnamed candidates (Lt. Governor, Land Commissioner, and Supreme Court Pl. 9) as in the two that did not involve Spanish-surnamed candidates (Supreme Court Pl. 3 and Supreme Court Pl. 5). Joint Expert Ex. E-2 at 27, Tables 2-5. He found that the polarization was not simply a function of partisanship, since it was often more stark within Democratic primaries with Latino and non-Latino candidates than it was in general elections. Id. at 26, 36. He noted that in the two statewide Democratic primary contests in 2010 that featured Spanish-surnamed candidates (Land Commis- sioner (Uribe) and Lt. Governor (Chavez-Thompson)), voting was more polarized along Latino/non-Latino lines than in the general election. Id. at 36 and Tables 7-11. Kousser did not find polarization within Republi- can primaries in 2010 because there were so few Latino voters. Id. at 36. 692. Kousser found racially polarized voting in general elections between 2002 and 2010 in which Spanish- Surnamed Republicans ran against Democrats in HD33, HD35, HD78, HD117, CD23, and CD27. Joint Expert Ex. E-2 at 36 and Tables 12-14. Using ordinary least squares and weighted least squares, Kousser found that African-American voters supported Latino Democratic but not Republican candidates in the general election almost unanimously. Id. at 51 and Tables 15-16. He concluded that, “[i]n general elections in Texas today, a black-Hispanic coalition in favor of Demo- cratic candidates is a fact of life.” Id. at 51. Kousser felt it made sense to combine African-American and Latino voters in districts to assess plans because they reliably coalesce in general elections. Joint Expert Ex. E-2 at 73, Tables 15-16. 469a 693. Lichtman noted the experts’ findings of racially polarized voting in the 2003 LULAC v. Perry litigation and confirmed the existence of racially polarized voting by performing ecological regression analysis on five general elections from 2008 and 201066 with a mix of candidates of different races. Joint Expert Ex. E-3 at 6-7. African-American voters had a mean of 99% of voters supporting the Democratic candidate; Latinos had a mean of 94%. Id. at 8 Table 2. In every county with the exception of Travis County, a substantial majority of Anglo voters voted against the candidates of choice of African-American and Latino voters. Joint Expert Ex. E-3 at 7. In Travis County, 52% of Anglo voters voted for the Democratic candidate in the general election. Id. at 8, Table 2. Statewide, the mean percentage of Anglo voters voting for the Democratic candidate was 29% including Travis County and 26% if Travis County is excluded. Id. Professor Lichtman opined that “[t]his polarized voting substantially impedes the ability of African-American and Latino voters to elect candidates of their choice in legislative districts for State House and Congress, given the substantially lower turnout of minorities than Anglos, which reflects both socio-economic disparities and histories of racial discrimination.” Joint Expert Ex. E- 3 at 9. In each of the five elections in the key counties he studied, Anglo turnout is higher than African- American turnout and substantially higher than Latino turnout. Id. at 9-10, Table 3. Because of the bloc voting, Lichtman opined that minority opportunity districts “must be carefully crafted to avoid the defeat

66 Professor Lichtman used the following five elections: 2008 President, 2008 U.S. Senate, 2008 Supreme Court Chief Justice, 2010 Governor, and 2010 Lt. Governor. Joint Expert Ex. E-3 at 8, Table 2. 470a of minority candidates of choice by Anglo bloc voting, especially given turnout differences across counties.” Id. at 9. 694. Dr. Ansolabahere found high levels of racial cohesion and polarized voting. Joint Expert Ex. E-15 at 7. In the average statewide and federal election, approximately 75% of Whites vote for the same candidate, while 75% or more of Hispanics and 90% or more of Blacks vote for the same candidate. The candidates chosen by a majority of Blacks and Hispanics are opposite to the candidates preferred by Whites. Id. Travis County is an exception to this pattern. Id. In Bexar, Tarrant, Dallas, Harris, and Nueces Counties, the voting patterns appear to be similarly racially polarized. Ansolabahere 8-22-11 depo. (Ex. J-44) at 31. And the rest of the state is similar to the non-Travis County urban areas. Id. 695. Murray’s homogeneous precinct analysis (using precincts greater than 85% BVAP, HVAP, or Anglo VAP) showed that general election voting in Texas is highly polarized along racial/ethnic lines, with size- able majorities of Anglo voters opposing the candidates favored by Black and Latino voters. Joint Expert Ex. E-4 at 20. In Harris County, Black voters supported Obama in 2008 at 98% and Hispanics at 63%, while Anglos supported McCain 73%. Joint Expert Ex. E-4 at 20. 696. Dr. Engstrom’s original report (Joint Expert Ex. E-7) looked at general elections for a variety of statewide offices in 2008 and 2010 in Bexar, Dallas, El Paso, Harris, Nueces, Tarrant, and Travis Counties, and a 52-county area in South Texas involving a Latino and non-Latino candidate. The 2010 general elections included: Lt. Governor (Chavez-Thompson (D)), Land Commissioner (Uribe (D)), and Supreme 471a Court Place 9 (Guzman (R)). The 2008 general elections included: U.S. Senate (Noriega (D)), Supreme Court Place 8 (Yanez (D)), and Court of Criminal Appeals Place 4 (Molina (D)). Using King’s ecological inference analysis, Engstrom conducted a bivariate analysis (Latino and non-Latino) for each area, and a multivariate analysis (Latinos, African Americans, and others) for the five largest counties (Bexar, Dallas, Harris, Tarrant, and Travis). He found that Latino voters were highly cohesive in support of Latino candidates with the Democratic Party nomination in all five general elections in all areas studied. Latino voters did not support Guzman, the only Republican Latino nominee in a general election (for Place 9 on the Supreme Court). The bivariate analysis showed that non-Latino voters were generally not supportive of Latino candidates and did not share the candidate preferences of Latino voters. However, the multivariate analysis showed that African-American voters did support the candidate preferences of Latino voters, while “other voters” (primarily Anglos) did not. In addition, there was more support for Latino candidates by non-Latino voters in Travis County than other areas. Other voters did provide support for the only Republican candidate that was a Latino in the general election (though only 50.11% in Travis County). 697. In his (corrected) Rebuttal Report (Joint Expert Ex. E-7; docket no. 307-1), Dr. Engstrom added the following 2006 general elections to his analysis: Lt. Governor (Alvarado (D), Supreme Court Place 4 (Medina (R)), Court of Criminal Appeals Presiding Judge (Molina (D)). He also made conclusions about racially polarized voting, finding the presence of racially polarized voting in general elections in El Paso County, Nueces County, South Texas, Dallas County, Harris County, Tarrant County, and Travis County. He again 472a found that Latino voters were very cohesive in their preferences for Latino candidates in general elections, but they do exercise discretion in choosing which Latino candidates to support. He found that Latino- preferred candidates were supported in general elections by African-American voters. Other voters (primarily Anglos) consistently cast a majority of their votes for the opponents of these candidates in general elections. Dr. Engstrom also found racially polarized voting in the 2010 general elections for CD23, CD27, HD33, and HD78, elections in which Latino-preferred incumbents in Latino opportunity districts were defeated. Dr. Engstrom assessed turnout differences between Latinos and non-Latinos in these elections and found the percentage of those turning out to vote that was Latino as follows: 40.77% in CD23; 46.72% in CD27; 45.08% in HD33; and 34.88% in HD78. The percentages of the overall votes received by the incumbents were: 44.44% in CD23; 47.11% in CD27; 47.49% in HD33; and 47.59% in HD78. 698. In his 2014 Supplemental Report on Racially Polarized Voting in Selected Areas of Texas (PL-967), Dr. Engstrom supplemented his prior report with 2012 statewide elections in which voters had a choice between a Latino and non-Latino candidate. The elections analyzed involved the 2012 Republican primary elections for U.S. Senate (Cruz) and Texas Supreme Court Place 4 (Medina) and the subsequent general elections for these seats. In the primary bivariate analysis, Cruz was the candidate of choice of Latinos generally (except in Tarrant County) and non-Latinos. But in the multivariate analysis, Cruz received a majority of votes from Latinos only in Dallas County, where he also received a majority of votes from African Americans. Other voters supported Cruz in all five counties analyzed. In the primary bivariate analysis, 473a Medina was the candidate of choice of Latino voters, but this preference was shared by non-Latinos in only Harris County, Tarrant County, and Travis County. In the multivariate analysis, the preference for Medina was shared by Latinos and “other voters” in only Harris County, while Latinos and African Americans shared support in Bexar, Dallas, and Harris Counties, and neither supported him in Tarrant and Travis Counties. Engstrom noted that “the most important estimates for assessing polarized voting in these counties and in the South Texas area are those for the General Election because much larger numbers of Latinos voted in this election than in the Republican nomination contests.” PL-967 at 5. Whereas 7,079 Latinos received ballots for the 2012 Republican runoff elections, 194,032 received ballots in the general election. Thus, the primary elections were much less probative of racially polarized voting than the general election. In the general election, none of the estimates in any area “come remotely close to indicating that Mr. Cruz is the candidate of choice for Latino voters.” For non-Latinos, Cruz received support in South Texas, Bexar County, El Paso County, Nueces County, and Tarrant County in the bivariate analysis. In the multivariate analysis, neither Latinos nor African Americans supported Cruz, but other voters supported him in Bexar, Dallas, Harris, and Tarrant Counties. 43.5% of other voters in Travis County supported Cruz. Engstrom wrote, “The estimates for Bexar, El Paso, Nueces, Tarrant, and South Texas reveal clear differences in candidate preferences between Latinos and non-Latino or non-minority voters. The only county in which these groups of voters preferred the same candidate according to both types of analyses is Travis, and that preference was for Mr. Sadler.” PL-967 at 6-7. Engstrom noted that the 2012 general election for 474a Senate was an exception to his earlier observation that Latinos consistently supported candidates from within their own group in general elections, and shows that Latinos exercise discretion in deciding which Latino candidates to support. It did not change his conclusion that racially polarized voting exists in these areas of Texas. PL-967 at 7. 699. Dr. Brischetto’s analysis of 2012 elections also found racially polarized voting in Nueces County. TrA935-43 (Brischetto). 700. Dr. Murray opined that the general election is the key test for racially polarized voting because these are the only partisan contests that reflect the racial and ethnic diversity of the state’s population. Joint Expert Ex. E-4 at 19. There are very few minorities participating in Republican primaries, and overall participation in Democratic primaries is heavily skewed to minority voters, except for Travis County and some upscale, inner city areas in Houston and Dallas. Murray opined that in Democratic primaries, the shrunken white vote is much more racially and ethni- cally tolerant of minority candidates than in past decades simply because Anglos who have problems supporting minority candidates either vote in the Republican primary or, more often, skip voting in the primary. Joint Expert Ex. E-4 at 19-21. 701. Dr. Murray’s supplemental report (docket no. 282-1) asserts that the “effective election” for Texas voters and candidates is the general election. Overall voting in general elections greatly exceeds combined primary voting. The fact that primaries have become much less important vis-à-vis general elections is reflected in the lack of meaningful contests in the March primaries. In Harris, Dallas, and Tarrant Counties, the turnout patterns, number of contested 475a elections, and comparative competitiveness in outcomes show that the general election is the critical contest. 702. Dr. Lichtman testified that the general election is the key election for examining cohesiveness and racially polarized voting because it is the election that sends the members to the Legislature and involves many more voters than primary elections. Tr1223-26, Tr1260. 703. Hispanic voters are politically cohesive in general and primary elections. 704. African-American voters are politically cohesive in general and primary elections. 705. Latinos and African Americans are often not cohesive within the Democratic primary. 706. African-American and Latino voters are politically cohesive in general elections in support of the Democratic candidate, regardless of the candidate’s race. 707. With the exception of Travis County, Anglo voters are politically cohesive in general elections in support of the Republican candidate, regardless of the candidate’s race. 708. The partisan divide along racial lines was reflected in the 2011 Texas Legislature, with Democrat members being 90% minority and Republican members being 98% Anglo. Joint Expert Ex. E-4 (Murray report) at 21. 709. The 2011 legislative session included a number of bills that exhibited anti-minority or anti-Hispanic sentiment or had potentially discriminatory effects on minorities, including voter ID, the early voting absentee bill, the Sanctuary Cities bill, and public 476a education funding. Tr810-12 (Turner); Tr890 (Murray); Joint Expert Ex. E-4 (Murray report) at 21; Veasey v. Perry, No. 13-CV-193 (S.D. Tex. Oct. 9, 2014) (docket no. 628). Dr. Flores opined that the 2011 legislative session was the most racially charged he had wit- nessed in his career, with Hispanics and Spanish speakers being principal targets. Tr436. Flores saw video of an individual trying to give testimony at a state transportation committee hearing in Spanish, and one of the committee members interrupted him and said it was an insult that he was speaking in Spanish. Tr436-38. Joint Expert Ex. E-8 (Flores report) at 6. 710. The Anglo Republican-dominated Legislature felt that their only obligation in terms of recognizing the minority growth was to draw new § 2 districts if the 50.1% CVAP threshold (or SSVR in the House map) was met by a single minority group and to not retrogress existing districts under § 5 of the VRA. Beyond that they felt no obligation to recognize minority growth by creating districts in which minority voters would have the ability to elect preferred candidates or to influence elections. A minority opportunity district or a minority coalition district based on African- American and Hispanic voters would likely be a Democratic-leaning district. TrA1922 (Alford); TrJ174 (Arrington). Therefore, any gains for VRA-protected minorities would come at the direct expense of the dominant Anglo Republican establishment in Texas, and the Anglo Republican establishment in the Texas Legislature had no interest in improving electoral opportunities for minority voters. Joint Expert Ex. E- 4 (Murray report) at 11. Republican legislators were hostile to the creation of any minority districts, and would provide minority opportunities only if they felt it was required by the VRA. 477a 711. Hispanics have a higher non-citizenship rate than Anglos and African Americans. Ansolabahere 8- 22-11 depo. (Ex. J-44) at 9. HCVAP as a percentage of HVAP varies in different parts of the State. Id. at 15. HVAP is not a reliable indicator for majority-HCVAP status or voting strength in many areas of Texas because of the high rate of Hispanic non-citizens. 712. For African-American districts and to measure African-American voting strength, mapdrawers relied on BVAP. TrA383 (Interiano). Mapdrawers knew that districts with 40% BVAP performed reliably for African-American voters. Interiano and others (Downton, Solomons, TLC) discussed the fact that the African- American districts were not 50% BCVAP and made a policy decision to maintain them. TrA339 (Interiano). Interiano treated Latino and African-American districts differently. For Latino districts, they were either majority or nothing, whereas an African-American district could be less than majority and still be protected. African-American districts could perform below 50% BVAP due to the presence of non-citizen Hispanics, and districts above 40% BVAP were treated as African-American districts rather than coalition districts. Tr1485-87 (Interiano). 713. Downton asserted that he was always looking for a 50% threshold for either African-American or Hispanic districts because he felt that Supreme Court cases set a 50% standard. Tr1012-13. He had heard that courts use a lower threshold for BVAP, and he did not know whether 50% BVAP was necessary for creating a performing African-American district. Tr1012-13 (Downton). He stated that they reviewed the House map to compare the number of districts with at least 40% BVAP, at least 45% BVAP, and at least 50% 478a BVAP, and they either stayed the same or went up. Tr1012. 714. With regard to §5 preclearance concerns, some legislators were told that the DOJ would look at population metrics of existing districts to determine a starting point, but they would also look at overall electoral reality, including whether a district is performing or effective, as opposed to just a pure mathematical analysis. D-590 at 81-82 (March 1, 2011 HRC hearing, Archer testimony). Hanna was familiar with the Red 225 reports and the 2001 DOJ letter that discussed election analysis as part of their evaluation. TrA1542-43 (Hanna). 715. The DOJ guidelines on § 5 state, “In determining whether the ability to elect exists in the benchmark plan and whether it continues in the proposed plan, the Attorney General does not rely on any pre- determined or fixed demographic percentages at any point in the assessment. Rather, in the Department’s view, this determination requires a functional analysis of the electoral behavior within the particular jurisdiction or election district.” Department of Jus- tice, Guidance Concerning Redistricting Under Sec- tion 5 of the Voting Rights Act, Federal Register, Vol. 76, No. 27, Wednesday, February 9, 2011, Notices, at 7471. It also states that “ability to elect either exists or it does not in any particular circumstance.” Id. 716. Dr. Murray testified that in determining whether a district will be effective, he would first look at total population, then VAP, and election results. Tr1039. Murray opined that whether an urban district will be effective for minority voters depends on the relative weights of the two minority groups. Tr1037. He testified that in urban Texas, any district that is much above 30% BVAP is likely to be effective, and at 479a 35% almost certainly the candidate supported by the African-American community gets elected. Tr1037. He testified that with Latino percentages, a lot depends on the other population in the district. An ideal Latino district could be as low as 50% Latino population if there is 15% African-American population in the same district, so the combined population is 65% or more. Id. In such a district, Latino voters would likely be successful in the primary, and in combination with African Americans in the general election (the effective election) would be able to support a candidate of their choice successfully. Id. In South Texas, a district would need a much higher SSVR than in urban areas like Dallas and Harris Counties to be effective for Latinos because there are only Anglos and Latinos and they do not vote for the same candidates. Tr1039 (Murray). 717. Dr. Kousser found that in 2008, Black and Hispanic voters were successful in electing their candidates of choice in all types of majority-minority districts he studied (HVAP-majority, HCVAP-majority, B+HVAP-majority, and BVAP+HCVAP-majority). Joint Expert Ex. E-2 at 74-76, Table 18. Kousser therefore opined that redistricting plans should be assessed not only by the number of HCVAP-majority districts they contain, but also combined majority districts. Joint Expert Ex. E-2 at 76. Although Black and Hispanic voters were less successful in 2010, B+HVAP-majority districts still provided four chances out of five of electing a minority-preferred candidate. Joint Expert Ex. E-2 at 76. Kousser opined that retrogression from the number of B+HVAP-majority seats would raise § 5 issues, and the failure to draw them would invite a § 2 challenge. Id. 480a 718. During redistricting, Hanna encouraged election analysis because there are shortcomings in the demographic analysis, especially as you get close to 50%. TrA1513 (Hanna). He stated that one would want to conduct election analysis to determine whether a district is performing or not, and for § 2 to determine whether the district truly provided an opportunity. However, he did not know whether opportunity meant success half the time, three quarters of the time, or some other number. TrA1513-14 (Hanna). Hanna did not think a district that was 0/10 would be a performing Latino opportunity district. TrA1567 (Hanna). He also likely would not consider 1/10 to be performing, and he did not advise anyone that 1/10 was performing. TrA1568 (Hanna). 719. Dr. Arrington testified that demographics alone are a misleading indicator of whether a district would perform or actually offer an opportunity; you have to look at performance. Tr432. 720. Downton did not think that effectiveness was an issue for § 2 compliance, and felt that if maps create the same number of opportunity districts (as defined solely by 50% HCVAP or SSVR), then it would be a political choice which configuration to choose. Downton 8-31-11 depo. (Ex. J-62) at 63. 721. Dr. Lichtman studied turnout rates for the various groups in Texas and the significance of turnout to minority ability to elect. Tr1226. He noted that opportunity to elect depends on three factors: demographic composition of the district, voting behavior of the groups, and turnout of the groups. Id. Turn- out is an important consideration in assessing electoral opportunities. Id. Dr. Lichtman opined that state policies with regard to economic development, education, welfare, and other matters can greatly affect 481a voter turnout, because turnout is very closely tied to socioeconomic standing. Tr1247. He also noted that the state can directly affect turnout with respect to laws such as voter registration and voter ID. Id. Thus, the state both directly and indirectly affects turnout. Id. Lichtman’s analysis showed that statewide there are turnout differences in general elections among the three groups, with Anglos participating at highest rates in terms of VAP, African Americans next highest, and Latinos the lowest. Tr1226-27. Overall the same patterns are observed at the county level, but turnout varies a great deal from county to county, most significantly among Hispanics. Tr1227. For example, in Bexar County estimated Latino turnout is 19%, in Tarrant County it is 4%, and in Harris County 6%. There is also county variation among African- American and Anglo voters, but it is not as large. Id. Dr. Lichtman opined that the turnout variability means that a single number or metric for measuring opportunity is not appropriate, and instead there must be a searching, practical inquiry that considers elections, not just numbers and demographics. Tr1227-28. Lichtman conducted an analysis of proposed alternative plans that looked at the number of majority- minority districts (in terms of VAP) and whether they were effective using reconstituted election analysis. Tr1228; Joint Expert Ex. E-3. He found that Plan H232 included eight additional effective minority House districts compared to the enacted plan, and Plan C121 had three additional effective minority congressional districts compared to Plan C185. Tr1231, Tr1241. 722. The OAG continuously provided information to Solomons’ redistricting staff, including the RPVA/ OAG 10. TrA1264-65 Solomons); TrA1563 (Hanna); Interiano 8-26-11 depo. (Ex. J-61) at 158. The staff used these analyses for determining whether a 482a proposed plan would comply with the VRA. TrA1265 (Solomons). Downton and Interiano understood that the OAG 10 indicated Hispanic performance in a district (not Democrat performance). Tr1456, TrA302- 04 (Interiano); Downton 8-12-11 depo. (Ex. J-62) at 58. Mapdrawers looked at the RPVA and OAG 10 to determine whether Hispanic candidates of choice were performing better or worse in plans under consideration. TrA1511-14 (Hanna); Interiano 8-26-11 depo. (Ex. J-61) at 89. Interiano and Downton used the OAG 10 to identify minority candidates of choice, to determine their performance in the benchmark and in districts under consideration, and to evaluate § 5 issues. TrA1637, TrA1695 (Downton); TrA1552 (Hanna); TrJ1615, TrA5, TrA356 (Interiano); TrJ1015-16 (Solomons). Solomons and his staff did not share the RPVA or OAG 10 with anybody (including minority legislators) on the HRC. TrA1266, TrA1325 (Solomons); TrA379 (Interiano); TrJ1958 (Bruce). 723. Legislators and mapdrawers had access to turnout data through the TLC. TrJ348 (Farias); D-578 Part 2 at 12 (Sept. 1, 2010 hearing transcript, Dyer testimony). Mapdrawers could shade geographic areas in RedAppl by percent turnout and view the number of votes cast in a particular VTD in a particular election. TrJ239-40 (Dyer). Mapdrawers did not have access to SSVR turnout data through the TLC. How- ever, RedAppl provided demographic data at the block level, and mapdrawers could identify precincts that were relatively high in Hispanic population but relatively low in total turnout. TrJ267 (Dyer). In addition, reports from the Texas OAG that mapdrawers Downton and Interiano received and relied upon included estimated turnout by race in specific districts. See US-3; D-182; D-183. 483a 724. Interiano was aware of Opiela’s nudge factor idea in November 2010, before he began working on any maps. US-75. Downton and Interiano also understood the concept of including more Hispanics in a minority opportunity district to increase its SSVR or HCVAP while simultaneouslyincluding areas with lower Hispanic turnout to protect Republican incumbents who were not the Hispanic candidate of choice. They were aware that the effectiveness of a minority district depended in large part on voter turnout and that manipulating Hispanic voter turnout would help protect Republican incumbents in minority districts. Downton 8-12-11 depo. (Ex. J-62) at 24-25; TrJ133 (Arrington). 725. Interiano tried to assist Opiela in getting data from TLC to allow him to implement the nudge factor. D-262; US-81; TrJ1482 (Interiano). Interiano obtained from TLC block level data showing SSVR/Total His- panic population with 2000 census blocks. TrJ291 (Dyer); D-262; US-185.67 This data would allow one to identify census blocks with a low turnout in the 2008 election and relatively high SSVR rates and thus to implement the nudge factor. Tr59 (Dyer); TrJ223 (Arrington). Further, the information that was available to mapdrawers using RedAppl would have allowed them to implement the nudge factor. TrJ130 (Arring- ton). Dyer tried to implement the nudge factor at the block level and it was “pretty slow.” TrJ287 (Dyer). It would be “very tedious” to do for a large district. Id.

67 The Census Bureau provides a block equivalency file, which Dyer testified was hard to work with, but she also testified that someone could compare the 2000 and 2010 census block information using the file. TrJ291 (Dyer). She did not recall if Interiano later asked for the information with updated 2010 census geography. TrJ294 (Dyer). 484a However, she also stated it would not be difficult, it just might take a while. TrJ296. 726. Both the House and congressional maps were developed in private by a few key players, with minimal public hearings or public input, and with minimal involvement from minority members. The plans were then voted out primarily along party lines. 727. There are approximately 8,400 precincts in Texas. Tr745 (Korbel); Joint Expert Ex. E-2 (Kousser report) at 11 n.7. 728. Splitting precincts is disruptive and tends to reduce voter turnout because of voter confusion. TrJ138-40 (Arrington). It also creates problems in terms of ballot design. TrJ138 (Arrington). It should therefore be avoided. TrJ139 (Arrington). It is especially disruptive for those people who find it most difficult to vote because of socioeconomic status. Id. Con- centrating precinct splits in minority communities weighs more heavily on minority voters and drives down minority voter turnout. Oct. 2011 Arrington report at 84-103. 729. Although there was no official policy against splitting precincts, Seliger agreed that not splitting precincts would be part of Texas’s traditional redistricting principles, though “not strictly adhered to.” TrA249. Solomons did not tell anyone that they could not split precincts “as an absolute type of rule” because “you may need to occasionally split a precinct.” TrJ1084. Rep. Pickett testified that Solomons asked members not to split precincts when drawing the House map. TrJ734. Downton testified that they asked the members not to split precincts because all of the individual map pieces would have to be put together like a puzzle. 485a TrJ2020.68 However, Downton did not try to limit or count precinct splits himself when he was drawing maps. TrJ2020, TrJ2024, TrJ2115 (Downton). He stated that he “didn’t have any prohibition or directive in my mind not to split them.” TrJ2020. 730. Plan C185 split 518 VTDs. US-699; PL-1633; TrA409-10 (Arrington); Tr690-691 (Korbel). CD35 has the most splits, with 106. Tr685 (Korbel). In Plan C185, the precinct splits are overwhelmingly located in the minority districts, and the difference between Anglo and minority districts is statistically significant TrA410 (Arrington). Arrington opined that the precinct splits were based on race and were evidence that race was being used as a proxy for partisanship. TrA437. Some of Plaintiffs’ proposed plans also have many precinct splits. TrA411 (Arrington) (C166 has 536; C163 has 447). Dr. Arrington did not examine alternative plans to determine whether the precinct splits in those plans were also concentrated in minority districts TrA434, TrA441 (Arrington). 731. Downton stated that he split precincts in the congressional map to equalize population and in his attempts to comply with the VRA, but he also split precincts at times when it was not required. TrJ2115, TrA1631 (Downton). Arrington agreed that equalizing population would require splitting some precincts, but not a lot, and there are race-neutral reasons why precincts may be split. TrA438-39 (Arrington). His analysis did not differentiate between racial and nonracial splits. TrA437-38. But he again felt it was unlikely that these race-neutral splits would occur

68 This would not be true as to drop-in counties in the Texas House map, however, because, by definition, those counties were “dropped in” and split precincts within the counties would not matter. 486a more in the minority districts or would split minority populations from majority populations. TrA440. 732. When mapdrawers split precincts in RedAppl, the only accurate data available below the precinct level was population and racial data. Because individual votes are unknown, political information was allocated homogeneously across the precincts. However, mapdrawers would see changes in political statistics when they split precincts because RedAppl allocated political data to the census block level. Split precincts did not accurately reflect political performance. 733. Texas has a long history of discriminating against its African-American and Hispanic residents, and there are lingering effects on voting and electoral participation from this past discrimination. Tr179-201 (Chapa); Joint Expert Ex. E-4 (Murray report) at 37; Joint Expert Ex. E-12 (Burton report) at 14. Texas had a white primary and a poll tax. TrJ160 (Arrington); Tr593 (Tijerina). Plaintiffs provided lay witness testimony from Sen. Gonzalo Barrientos, Sen. Joe Bernal, Alex Jimenez, Congresswoman Eddie Bernice Johnson, and Howard Jefferson about past racial discrimination they experienced. 734. In Texas redistricting, there have been consistent attempts to minimize the opportunity of Latino communities to elect the candidates of their choice. Tr435 (Flores). Texas and its political subdivisions have had over 200 voting rights challenges since 1982. Joint Expert Ex. E-8 (Flores report) at 5. Dr. Arrington testified that in every decade since 1970 Texas has passed one or more redistricting plans after the census that have been declared either unconstitutional or violations of the VRA. TrJ158. Rep. Anchia testified that Latinos continue to face obstacles in voting through 487a lack of same day registration, voter ID requirements, and language barriers. Anchia depo. at 72. 735. Dr. Chapa testified that Latinos have experienced discrimination that results in the lingering effects of lower income and education levels and lower participation rates in voting, registration, turnout, and politics in general compared to non-Hispanics in Texas. Tr189; Joint Expert Ex. E-1 at 4-5, 16. Dr. Chapa’s report was admitted without objection, and therein he established that Texas Hispanics have lower levels of both education and income when compared to non-Hispanics, and these disparities have persisted throughout the 20th Century. He noted that there have been a plethora of studies showing that people with lower levels of education and lower incomes participate less in the electoral system, and noted that lingering effects of past discrimination against Latinos in Texas result in lower Latino participation rates in voting and registration and politics in general. Tr189, Tr200. 736. Dr. Andres Tijerina submitted a detailed report on the history of the violation of civil rights of Latinos in Texas with emphasis on the electoral process and voting, which was admitted without objection. Joint Expert Ex. E-10. He found that discrimination against Latinos in Texas has been a pervasive and constant phenomenon since 1836, when Anglo-Americans took control of Texas government. He concluded that the legacy of 150 years of multi-faceted government- condoned discrimination against Mexican Americans is a state educational system that maintains a high drop out rate and is still characterized by widespread segregation. He concluded that Mexican Americans in Texas still bear the effects of historical discrimination, which hinders their ability to participate effectively in 488a the political process and that it “is clear that the lower rates of voter registration, voting, and running for elective office are directly related to this discrimination.” Id. at 32. He testified that the past discrimination affects voter registration and turnout rates, as well as the ability to run for office or organize a campaign. Tr595-96. 737. Dr. Orville Vernon Burton also discussed Texas’s history of racial discrimination in his report, including in voting. Joint Expert Ex. E-12. He noted that socioeconomic factors have a direct influence on voting, with lower socioeconomic status resulting in decreased political participation. Id. at 36. He concluded that the “socioeconomic disadvantages experienced by African Americans and Latinos constitute a clear hindrance to the effective participation of these groups in the political process.” Id. He notes that in Texas, Latinos and African Americans have higher unemployment rates than Anglos and lag in education. He noted that home ownership and residence influence voting behavior, and related issues have negatively affected minority political participation. Id. at 45-47. He also found disparities in health and access to insurance and health care. Id. at 47-48. Dr. Burton noted that voter turnout depends on many factors, and that socioeconomic differences between the races are one cause of differences in participation rates. He opined that a history of de jure segregation and present day effects of past racial discrimination may also account for some of the turnout differences, as well as lack of transportation or literacy issues. Minority groups are more likely to participate if they have some reasonable hope of electing their candidate of choice, while districts drawn to split minority communities and ensuring defeat of minority-preferred candidates have a chilling effect. Id. at 52-53. 489a 738. Dr. Gonzalez-Baker’s report was admitted without objection. She concluded that it is clear that Latino educational achievement lags far behind that of non-Latino Whites in the same age bracket. Joint Expert Ex. E-9 at 7. Latinos continue to display pro- found evidence of social and economic disadvantage around the state. Gonzalez-Baker depo. (Ex. J-41) at 64. 739. Education and income play a big role in voting. TrJ1387 (Korbel). Per Capita Income (based on 2008- 2012 ACS data) for Hispanics is $14,169, for African- Americans is $18,418, and for Whites is $34,826. TrJ1387 (Korbel); MALC-139. African Americans and Hispanics are approximately four times more likely than Whites to be in poverty. TrJ1388 (Korbel); MALC- 139. Unemployment rates are higher for African Americans and Hispanics than Whites. TrJ1389 (Korbel); MALC-139. 25.4% of Hispanics are functionally illit- erate, compared to 2.4% of Whites. TrJ1389 (Korbel); MALC-139. Almost 42% of Hispanics over the age of 25 have not graduated from high school, compared to 8.7% of whites. TrJ1389 (Korbel); MALC-139. These trends and the differences between minority and white population have remained fairly constant over the years. TrJ1390-91 (Korbel). The data before the Legislature would be virtually the same. TrJ1391 (Korbel). 740. Dr. Engstrom noted that both Republicans and Democrats have discriminated against minorities. TrA531. He felt that both parties are unsure of how Latinos will vote and so try to control their vote. TrA532-33. Rep. Senfronia Thompson agreed that both Democrats and Republicans have engaged in voting rights violations that had to be remedied in court. TrJ1228. 490a 741. Historically, minorities in Texas have been severely under-represented in the U.S. Congress. Joint Expert Ex. E-12 (Burton report) at 54. Texas did not send its first Latino representative to Congress until 1961 (Henry B. Gonzalez). The first African American to serve in Congress was Barbara Jordan, who was elected after enactment of the VRA. Id. at 55. In 2010, Anglos were less than 50% of the total population of Texas but held 21 of 32 congressional seats (11 seats were represented by minorities, eight Latinos and three African Americans). Id. Historically, there have been only six African-American congresspersons from Texas in the U.S. House. Id. at 56. In the Texas Legislature in 2011, Hispanics held 21.1% of seats and African Americans held 11.1% of seats. Rep. Thompson testified that there were eight African-American members in the Texas House in the 2013 session. TrJ1226. 742. Anglo Republican congresspersons have not been responsive to the African-American community. Congressman Roger Williams CD25 and Congressman Blake Farenthold both receive an “F” on the NAACP report card. TrA701. The NAACP report card is an effective barometer for determining which members of Congress support the interests of the African-American community. Tr1297 (Johnson); Tr1386 (Jefferson). Some Democrat Latino congresspersons have effectively supported the NAACP agenda. Tr1388-90 (Jefferson). SIGNED on this 10th day of March, 2017. /s/ XAVIER RODRIGUEZ UNITED STATES DISTRICT JUDGE /s/ ORLANDO L. GARCIA CHIEF UNITED STATES DISTRICT JUDGE